TW202205300A - Methylation pattern analysis of haplotypes in tissues in a dna mixture - Google Patents

Methylation pattern analysis of haplotypes in tissues in a dna mixture Download PDF

Info

Publication number
TW202205300A
TW202205300A TW110121521A TW110121521A TW202205300A TW 202205300 A TW202205300 A TW 202205300A TW 110121521 A TW110121521 A TW 110121521A TW 110121521 A TW110121521 A TW 110121521A TW 202205300 A TW202205300 A TW 202205300A
Authority
TW
Taiwan
Prior art keywords
methylation
haplotype
free dna
dna molecules
tissue
Prior art date
Application number
TW110121521A
Other languages
Chinese (zh)
Other versions
TWI798718B (en
Inventor
煜明 盧
君賜 陳
慧君 趙
江培勇
孫坤
Original Assignee
香港中文大學
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 香港中文大學 filed Critical 香港中文大學
Publication of TW202205300A publication Critical patent/TW202205300A/en
Application granted granted Critical
Publication of TWI798718B publication Critical patent/TWI798718B/en

Links

Images

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B30/00—ICT specially adapted for sequence analysis involving nucleotides or amino acids
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/154—Methylation markers
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/156—Polymorphic or mutational markers
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/172—Haplotypes

Landscapes

  • Chemical & Material Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Organic Chemistry (AREA)
  • Engineering & Computer Science (AREA)
  • Analytical Chemistry (AREA)
  • Genetics & Genomics (AREA)
  • Wood Science & Technology (AREA)
  • Zoology (AREA)
  • Physics & Mathematics (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Biophysics (AREA)
  • Biotechnology (AREA)
  • General Health & Medical Sciences (AREA)
  • Microbiology (AREA)
  • Molecular Biology (AREA)
  • Biochemistry (AREA)
  • General Engineering & Computer Science (AREA)
  • Oncology (AREA)
  • Hospice & Palliative Care (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Medical Informatics (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)

Abstract

Systems, apparatuses, and method are provided for determining the contributions of different tissues to a biological sample that includes a mixture of cell-free DNA molecules from various tissues types, e.g., as occurs in plasma or serum and other body fluids. Embodiments can analyze the methylation patterns of the DNA mixture (e.g., methylation levels at particular loci) for a particular haplotype and determine fractional contributions of various tissue types to the DNA mixture, e.g., of fetal tissue types or tissue types of specific organs that might have a tumor. Such fractional contributions determined for a haplotype can be used in a variety of ways.

Description

DNA混合物中組織之單倍型甲基化模式分析Haplotype methylation pattern analysis of tissues in DNA mixtures

先前已展示經由對懷有胎兒之懷孕女性之血漿DNA之分析，可使用相對單倍型劑量分析(RHDO)之方法推論由胎兒遺傳之母體單倍型(Lo等人 Sci Transl Med 2010; 2: 61ra91及美國專利8,467,976)。可使用關於懷孕女性之單倍型資訊。可使用家族分析或直接分析單倍型之方法(例如Fan等人 Nat Biotechnol 2011; 29: 51-57；Snyder等人 Nat Rev Genet 2015; 16: 344-358)獲得單倍型資訊。在母親中雜合但在父親中純合之SNP可用於RHDO分析。 此類使用特定SNP可限制可使用的基因座，且因此限制資料及精確性的量。此類使用特定SNP亦可限制方法之臨床效用，因為來自額外家族成員之DNA樣品可能不可用，且直接分析單倍型之方法將增加分析成本。It has been previously shown that maternal haplotypes inherited from the fetus can be inferred using the method of relative haplotype dose analysis (RHDO) from the analysis of plasma DNA of pregnant women carrying the fetus (Lo et al. Sci Transl Med 2010; 2: 61ra91 and US Patent 8,467,976). Haplotype information on pregnant women is available. Haplotype information can be obtained using methods of family analysis or direct analysis of haplotypes (eg, Fan et al. Nat Biotechnol 2011; 29: 51-57; Snyder et al. Nat Rev Genet 2015; 16: 344-358). SNPs that are heterozygous in the mother but homozygous in the father can be used for RHDO analysis. Such use of specific SNPs can limit the loci that can be used, and thus limit the amount of data and precision. Such use of specific SNPs can also limit the clinical utility of the method, as DNA samples from additional family members may not be available, and methods to directly analyze haplotypes would increase the cost of analysis.

描述用於測定不同組織對包括來自各種組織類型之游離DNA分子之混合物之生物樣品之貢獻(例如如在血漿或血清及其他體液中進行)之實施例。實施例可對於特定單倍型分析DNA混合物之甲基化模式(例如特定基因座之甲基化程度)且測定各種組織類型，例如胚胎組織類型或可具有腫瘤的特定器官的組織類型對DNA混合物的百分比貢獻。對於單倍型測定的此類百分比貢獻可以多種方式使用。 在一些實施例中，可使用兩組來自母體樣品之游離DNA分子之甲基化程度測定組織類型之兩個百分比貢獻，各組用於胎兒之親體之兩個親體單倍型中的不同一者，用於分析之染色體區域。在各種實施方案中，母體樣品可為來自懷有一或多個胎兒之女性的血漿或血清樣品。兩個百分比貢獻可用於鑑別胚胎基因組的一部分。舉例而言，胚胎組織之兩個百分比貢獻之間的分離值可指示基因座之胚胎基因型且可指示兩個親本單倍型中之何者由胎兒遺傳。舉例而言，較高百分比貢獻可指示遺傳之單倍型，且若分離值小於臨限值，則可遺傳兩個單倍型；當兩個親體對於分析區域共用單倍型(或關於基因型之等位基因)時，可遺傳兩個單倍型。 在一些實施例中，對於一個單倍型，僅測定胚胎組織之一個百分比貢獻。當一個百分比貢獻超過參考值(例如由其他樣品測定)時，胎兒可測定為對於分析區域遺傳該單倍型。 在一些實施例中，可對於兩組來自母體樣品的游離DNA分子測定兩個甲基化程度，各組用於胎兒之親體之兩個親體單倍型中的不同一者，作為鑑別一部分胚胎基因組之一部分。兩個甲基化程度可彼此比較以鑑別由胎兒遺傳何種單倍型，例如藉由哪個甲基化程度較低。舉例而言，胎兒貢獻低甲基化的游離DNA分子，且一種單倍型之較低甲基化程度之量測結果指示該單倍型由胎兒遺傳。 在一些實施例中，可使用來自複數個組織類型之游離DNA分子之混合物偵測胎兒之目標染色體區域之序列不平衡。可鑑別具有第一目標單倍型及第二目標單倍型的目標染色體區域之目標雜合基因座，該第一目標單倍型與該第二目標單倍型具有不同等位基因。混合物中之胚胎組織類型之第一目標百分比貢獻可使用靶雜合基因座處之甲基化程度測定，其中甲基化程度使用位於(亦即覆蓋)第一單倍型之基因座處之游離DNA分子之目標組測定。類似地，可測定胚胎組織類型之第一參考百分比貢獻。可比較第一目標百分比貢獻及第一參考百分比貢獻之分離值與臨限值以確定胎兒是否具有序列不平衡。若兩個百分比貢獻顯著不同，則可確定序列不平衡。使用之特定臨限值可取決於測試之特定序列不平衡(例如擴增或缺失)。 在一些實施例中，第一組織類型中之第一單倍型之百分比貢獻可用於確定第一組織類型是否具有疾病病況。第一單倍型可具有對健康細胞或異常細胞具有特異性之簽名。因此，第一單倍型可不存在於生物體之健康細胞中，或存在於生物體之健康細胞中且不存在於可在混合物中之異常細胞中。可比較第一百分比貢獻與參考百分比貢獻之間的分離值與臨限值以確定第一組織類型是否具有疾病病況之分類。 在一些實施例中，拷貝數變異之組織來源可使用甲基化解卷積測定。第一染色體區域可鑑別為展現拷貝數變異。可測定第一染色體區域中之單倍型。對於M個組織類型中之每一者，可測定對應第一百分比貢獻與對應第二百分比貢獻之間的對應分離值。具有最高分離值之組織類型可鑑別為源組織。 其他實施例係關於與本文所描述之方法相關聯的系統及電腦可讀媒體。 可參考以下詳細描述及隨附圖式來獲得對本發明實施例之性質及優勢的較佳理解。Embodiments are described for determining the contribution of different tissues to biological samples comprising mixtures of cell-free DNA molecules from various tissue types (eg, as performed in plasma or serum and other body fluids). Embodiments can analyze methylation patterns of DNA mixtures for specific haplotypes (eg, the degree of methylation at specific loci) and determine various tissue types, such as embryonic tissue types or tissue types of specific organs that can have tumors versus DNA mixtures percentage contribution. Such percentage contributions to haplotype determination can be used in a variety of ways. In some embodiments, the degree of methylation of cell-free DNA molecules from a maternal sample can be used to determine the two percent contributions of tissue types, each for a different one of the two parental haplotypes of the parent of the fetus , the chromosomal region used for the analysis. In various embodiments, the maternal sample can be a plasma or serum sample from a woman pregnant with one or more fetuses. Two percent contributions can be used to identify a portion of an embryo's genome. For example, the separation value between two percent contributions of embryonic tissue can indicate the embryonic genotype of the locus and can indicate which of the two parental haplotypes is inherited by the fetus. For example, a higher percentage contribution can indicate an inherited haplotype, and if the segregation value is less than a threshold value, two haplotypes can be inherited; alleles), two haplotypes can be inherited. In some embodiments, only one percent contribution of embryonic tissue is determined for a haplotype. When a percentage contribution exceeds a reference value (eg, as determined from other samples), the fetus can be determined to have inherited that haplotype for the region analyzed. In some embodiments, two methylation levels can be determined for two sets of cell-free DNA molecules from a maternal sample, each set for a different one of the two parental haplotypes of the parent of the fetus, as part of the identification of the embryonic genome one part. The two methylation levels can be compared to each other to identify which haplotype is inherited by the fetus, eg, by which methylation level is lower. For example, the fetus contributes hypomethylated cell-free DNA molecules, and a measurement of the lower degree of methylation of a haplotype indicates that the haplotype is inherited by the fetus. In some embodiments, a mixture of cell-free DNA molecules from multiple tissue types can be used to detect sequence imbalances in target chromosomal regions of the fetus. A heterozygous locus of interest for a chromosomal region of interest having a first haplotype of interest and a second haplotype of interest that have different alleles can be identified. The first target percent contribution of the embryonic tissue types in the mixture can be determined using the degree of methylation at the target heterozygous locus, where the degree of methylation is determined using the degree of methylation at the locus located (ie, covering) the first haplotype. Target group determination of DNA molecules. Similarly, the first reference percent contribution of the embryonic tissue type can be determined. The separation value of the first target percent contribution and the first reference percent contribution can be compared to a threshold value to determine whether the fetus has sequence imbalance. Sequence imbalance can be determined if the two percent contributions are significantly different. The particular threshold used may depend on the particular sequence imbalance (eg, amplification or deletion) being tested. In some embodiments, the percent contribution of the first haplotype in the first tissue type can be used to determine whether the first tissue type has a disease condition. The first haplotype may have a signature specific for healthy cells or abnormal cells. Thus, the first haplotype may not be present in healthy cells of the organism, or may be present in healthy cells of the organism and absent in abnormal cells that may be in the mixture. The separation value between the first percent contribution and the reference percent contribution can be compared to a threshold value to determine whether the first tissue type has a classification of a disease condition. In some embodiments, the tissue source of copy number variation can be determined using methylation deconvolution. The first chromosomal region can be identified as exhibiting copy number variation. Haplotypes in the first chromosomal region can be determined. For each of the M tissue types, a corresponding separation value between the corresponding first percent contribution and the corresponding second percent contribution may be determined. The tissue type with the highest separation value can be identified as the source tissue. Other embodiments relate to systems and computer-readable media associated with the methods described herein. A better understanding of the nature and advantages of embodiments of the present invention can be obtained by reference to the following detailed description and accompanying drawings.

[相關申請案之交叉參考] 本申請案主張2015年7月20日申請之名稱為「Methylation Pattern Analysis Of Haplotypes In Tissues In A DNA Mixture」之美國臨時申請案第62/194,702號之優先權，其全部內容出於所有目的以引用的方式併入本文中。 術語 「甲基化組」提供基因組中複數個位點或基因座之DNA甲基化量的量度。甲基化組可對應於所有基因組、基因組之大部分或基因組之相對較小部分。所關注的甲基化組之實例為可將DNA貢獻至體液(例如血漿、血清、汗液、唾液、尿液、生殖器分泌物、精液、糞便流體、腹瀉流體、腦脊髓液、胃腸道分泌物、腹水流體、胸膜液、眼內流體、來自水囊腫(例如睾丸)之流體、來自包囊之流體、胰臟分泌物、腸道分泌物、痰、眼淚、來自乳房及甲狀腺之抽吸流體等)中之器官之甲基化組(例如腦細胞、骨骼、肺、心臟、肌肉及腎等之甲基化組)。器官可為移植器官。胎兒之甲基化組為另一實例。 「血漿甲基化組」為自動物(例如人類)之血漿或血清測定之甲基化組。血漿甲基化組為游離甲基化組之一個實例，因為血漿及血清包括游離DNA。血漿甲基化組亦為混合甲基化組之實例，因為其為胚胎/母體甲基化組或腫瘤/患者甲基化組或衍生自背景或器官移植中之不同組織或器官或供體/受體甲基化組之DNA的混合物。 「位點」(亦稱作「基因組位點」)對應於單一位點，其可為單一鹼基位置或相關鹼基位置群，例如CpG位點或相關鹼基位置之較大群。「基因座」可對應於包括多個位點之區域。基因座可僅包括一個位點，此將使得基因座在該背景下等效於一個位點。 各基因組位點(例如CpG位點)之「甲基化指數」可指在該位點顯示甲基化之DNA片段(例如如測定自序列讀數或探針)相對於覆蓋該位點之讀數總數之比例。「讀數」可對應於獲自DNA片段之資訊(例如位點之甲基化狀態)。讀數可使用優先雜交至特定甲基化狀態之DNA片段之試劑(例如引物或探針)獲得。通常，此類試劑在藉由取決於DNA分子之甲基化狀態不同地修飾或不同地識別DNA分子之方法，例如亞硫酸氫鹽轉化，或甲基化敏感限制酶，或甲基化結合蛋白，或抗甲基胞嘧啶抗體，或識別甲基胞嘧啶及羥甲基胞嘧啶之單分子定序技術處理後施用。 區域之「甲基化密度」可指顯示甲基化之區域內之位點之讀數數目除以覆蓋區域中之位點之讀數總數。位點可具有特定特徵，例如為CpG位點。因此，區域之「CpG甲基化密度」可指顯示CpG甲基化之讀數數目除以覆蓋區域中之CpG位點(例如特定CpG位點、CpG島或較大區域內之CpG位點)之讀數總數。舉例而言，人類基因組中每100 kb分區(bin)之甲基化密度可自亞硫酸氫鹽處理之後於CpG位點未轉化之胞嘧啶(其對應於甲基化胞嘧啶)的總數測定為映射至100 kb區域之序列讀數所覆蓋之所有CpG位點的比例。此分析亦可對於其他分區大小，例如500 bp、5 kb、10 kb、50 kb或1 Mb等進行。區域可為整個基因組或染色體或染色體之一部分(例如染色體臂)。當區域僅僅包括CpG位點時，CpG位點之甲基化指數與區域之甲基化密度相同。「甲基化胞嘧啶之比例」可指相比於分析之胞嘧啶殘基總數展示為甲基化(例如在亞硫酸氫鹽轉化之後未經轉化)之胞嘧啶位點「C」之數目，即包括區域中除CpG背景之外的胞嘧啶。甲基化指數、甲基化密度及甲基化胞嘧啶之比例為「甲基化程度」之實例。除亞硫酸氫鹽轉化之外，熟習此項技術者已知之其他方法可用於查詢DNA分子之甲基化狀態，包括(但不限於)對甲基化狀態敏感之酶(例如甲基化敏感限制酶)、甲基化結合蛋白、使用對甲基化狀態敏感之平台之單分子定序(例如奈米孔定序(Schreiber等人 Proc Natl Acad Sci 2013; 110: 18910-18915)及藉由Pacific Biosciences單分子實時分析(Flusberg等人 Nat Methods 2010; 7: 461-465))。 「甲基化圖譜」(亦稱為甲基化狀態)包括與區域之DNA甲基化有關之資訊。與DNA甲基化有關之資訊可包括(但不限於)CpG位點之甲基化指數、區域中CpG位點之甲基化密度、相鄰區域上CpG位點的分佈、含有超過一個CpG位點之區域內各個別CpG位點之甲基化模式或程度以及非CpG甲基化。基因組之大部分甲基化圖譜可視為等效於甲基化組。哺乳動物基因組中之「DNA甲基化」通常係指添加甲基至CpG二核苷酸中之胞嘧啶殘基的5'碳(即5-甲基胞嘧啶)。DNA甲基化可在例如CHG和CHH之其他情況下發生於胞嘧啶中，其中H為腺嘌呤、胞嘧啶或胸腺嘧啶。胞嘧啶甲基化亦可呈5-羥甲基胞嘧啶形式。亦報導非胞嘧啶甲基化，如N6 -甲基腺嘌呤。 「甲基化感測定序」係指允許吾人在定序方法期間確定DNA分子之甲基化狀態的任何定序方法，包括(但不限於)亞硫酸氫鹽定序、或定序前甲基化敏感限制酶消化、使用抗-甲基胞嘧啶抗體或甲基化結合蛋白之免疫沈澱或可說明甲基化狀態之單分子定序。 「組織」對應於一群細胞，其共同歸類為一個功能單元。單一組織中可發現超過一種類型的細胞。不同類型之組織可由不同類型之細胞(例如肝細胞、肺泡細胞或血細胞)組成，但亦可對應於來自不同生物體(母親與胎兒)的組織或對應於健康細胞與腫瘤細胞。「參考組織」對應於用於測定組織特異性甲基化程度之組織。來自不同個體之相同組織類型之多個樣品可用於測定該組織類型之組織特異性甲基化程度。 「生物樣品」係指獲自個體(例如人類，如懷孕女性、患有癌症之個體或疑似患有癌症之個體、器官移植受體或疑似具有涉及器官(例如心肌梗塞中之心臟、或中風中之大腦、或貧血中之造血系統)之疾病過程之個體)且含有一或多個所關注的核酸分子之任何樣品。生物樣品可為體液，如血液、血漿、血清、尿液、陰道流體、來自水囊腫(例如睾丸)之流體、或陰道沖洗流體、胸膜液、腹水流體、腦脊髓液、唾液、汗液、淚液、痰、支氣管肺泡灌洗術流體等。亦可使用糞便樣品。在各種實施例中，游離DNA已增濃之生物樣品(例如經由離心方案獲得的血漿樣品)中之大部分DNA可游離(與細胞相對)，例如大於50%、60%、70%、80%、90%、95%或99%。離心方案可包括3,000 g×10分鐘、獲得流體部分及再在30,000 g下再離心10分鐘以移除殘餘細胞。 術語「癌症程度」可指是否存在癌症(亦即存在或不存在)、癌症階段、腫瘤尺寸、是否存在轉移、身體總腫瘤負荷，及/或癌症嚴重度之其他量度(例如癌症復發)。癌症程度可為數字或其他標誌，諸如符號、字母表字母及顏色。程度可為零。癌症程度亦包括與突變或多種突變相關的癌變前或癌前期病狀(狀態)。可以各種方式使用癌症程度。舉例而言，篩選可檢查已知先前未患癌症之某人是否存在癌症。評估可調查已診斷患有癌症之某人以監測癌症隨時間之進展，研究治療有效性或確定預後。在一個實施例中，預後可用患者死於癌症之機率或特定期限或時間之後癌症進展之機率或癌症轉移之機率表示。偵測可意謂『篩選』或可意謂檢查暗示有癌症特徵(例如症狀或其他陽性測試)的某人是否患有癌症。 術語染色體區域之「序列不平衡」可在生物體健康之情況下指呈相對於期望值之來自染色體區域之游離之DNA分子之量的任何顯著偏差。舉例而言，染色體區域可展現某一組織中之擴增或缺失，進而導致DNA混合物(含有與來自其他組織之DNA混合之來自組織之DNA)中之染色體區域之序列不平衡。舉例而言，期望值可獲自另一樣品或獲自假設正常之另一染色體區域(例如代表二倍體生物體之兩個複本之量)。染色體區域可由多個不相交子區組成。 基因組基因座(標記物)之「類型」對應於基因座跨越組織類型之特定屬性。說明書主要係指I型基因座及II型基因座，其特性詳細提供於下文。給定類型之基因座可具有跨越組織類型之甲基化程度之特定統計變化。基因組基因座(標記物)之「類別」對應於跨越相同組織類型之不同個體之基因座之甲基化程度之特定變化。一組基因座(標記物)可由任何數目的各種類型及/或類別之基因座組成。因此，一組基因座對應於針對特定量測值選擇之基因座且不暗示組中之基因座之任何特定特性。 「分離值」對應於涉及兩個值，例如兩個百分比貢獻或兩個甲基化程度之差值或比率。分離值可為簡單差值或比率。分離值可包括其他係數，例如相乘係數。作為其他實例，可使用該等值之函數的差值或比率，例如兩個值之自然對數(ln)的差值或比率。分離值可包括差值及比率。 如本文所用，術語「分類 」係指與樣品之特定特性相關之任何數字或其他字符。舉例而言，符號「+」(或字語「正性」)可表示樣品歸類為具有缺失或擴增。分類可為二元(例如正性或負性)或具有更多分類層級(例如1至10或0至1之標度)。術語「閾值 」及「臨限值 」係指使用於操作之預定數字。臨限值可為高於或低於特定分類適用之值。可在此等上下文中之任一者中使用此等術語中之任一者。 DNA混合物(例如血漿)中之組織類型(例如胚胎組織、肝臟等)之間的甲基化差異可用於區分特定組織類型組織單倍型之特性。舉例而言，懷孕女性之血漿中之兩個母體單倍型之甲基化程度可用於確定何種單倍型自母親遺傳至胎兒。作為另一實例，胚胎組織中之兩個單倍型之甲基化程度可用於偵測胎兒中之序列不平衡(例如非整倍性)。亦可分析其他組織類型，例如用以偵測特定組織類型中之疾病病況。亦可測定拷貝數變異所發源之組織類型。 一些實施例可使用特定組織類型之某些基因組位點之已知甲基化程度自各種組織類型測定血漿(或其他DNA混合物)中之游離DNA之百分比。舉例而言，可對於肝臟樣品量測基因組位點之甲基化程度，且此等組織特異性甲基化程度可用於測定混合物中之多少游離DNA來自肝臟。可量測對DNA混合物提供重大貢獻之組織類型之甲基化程度，以使得可計算游離DNA混合物之優勢(例如大於90%、95%或99%)。此類其他樣品可包括(但不限於)以下中的一些或全部：肺、結腸、小腸、胰臟、腎上腺、食道、脂肪組織、心臟及大腦。 解卷積方法可用於測定已知組織特異性甲基化程度之組織類型中之每一者之百分比貢獻(例如百分比)。在一些實施例中，線性方程組可自已知組織特異性甲基化程度及指定基因組位點之混合物甲基化程度產生，且可測定(例如使用最小平方)最佳近似量測之混合物甲基化程度之百分比貢獻。 一旦測定百分比貢獻，百分比貢獻可用於各種目的。舉例而言，胚胎組織之百分比貢獻差異可用於測定自親體遺傳何種單倍型。可對於兩個親體單倍型中之每一者測定一或多個雜合基因座處之等位基因。一或多個雜合基因座處之游離DNA可用於測定兩個百分比貢獻：各單倍型各一個。舉例而言，具有第一單倍型之等位基因之游離之DNA分子可用於測定第一百分比貢獻，且具有第二單倍型之等位基因之游離之DNA分子可用於測定第二百分比貢獻。遺傳之單倍型將對應於胚胎組織之較高百分比貢獻。 另外，遺傳之單倍型將由於胚胎游離DNA之一般低甲基化而具有較低甲基化程度。可比較兩個單倍型之甲基化程度，且具有較低甲基化程度之單倍型可鑑別為遺傳之單倍型。 作為另一實例，可在胎兒之目標染色體區域中偵測到序列不平衡。可測定目標染色體區域中之第一單倍型之混合物中之胚胎組織類型之目標百分比貢獻。類似地，可測定參考染色體區域之胚胎組織類型之參考百分比貢獻。兩個貢獻之間的分離值可相比於臨限值以確定胎兒是否具有序列不平衡(例如非整倍性)。 作為另一實例，第一單倍型可具有特異於健康細胞或異常細胞之簽名。針對第一單倍型測定之百分比貢獻於參考百分比貢獻之間的分離值可相比於臨限值以確定第一組織類型是否具有疾病病況之分類。作為實例，第一單倍型可為移植器官或腫瘤，或僅在健康細胞中且不再移植器官或腫瘤中。疾病病況可為移植器官是否經排斥，或腫瘤是否尺寸增加或轉移(例如在不移除所有腫瘤之手術之後)。 作為另一實例，拷貝數變異之組織來源可使用甲基化解卷積測定。第一染色體區域可鑑別為展現拷貝數變異。對於M個組織類型中之每一者，可測定第一染色體區域中之兩個單倍型之百分比貢獻之間的對應分離值。具有最高分離值之組織類型可鑑別為源組織。 首先描述甲基化解卷積，且接著描述甲基化標記之選擇及甲基化解卷積之精確性。接著描述百分比貢獻用於測定胚胎基因組部分之用途。 I.根據甲基化解卷積之DNA混合物之組成 不同組織類型可對於基因組位點具有不同甲基化程度。此等差異可用於測定混合物中來自各種組織類型之DNA之百分比貢獻。因此，可藉由組織特異性甲基化模式分析測定DNA混合物之組成。以下實例論述甲基化密度，但可使用其他甲基化程度。A . 單一基因組位點 甲基化解卷積之原理可使用單一甲基化基因組位點(甲基化標記物)測定來自生物體之DNA混合物之組成而說明。假定組織A對於基因組位點完全甲基化，即100%之甲基化密度(MD)且組織B完全未甲基化，即0%之MD。在此實例中，甲基化密度係指CpG二核苷酸在所關注的區域中經甲基化之背景下之胞嘧啶殘基之百分比。 若DNA混合物C由組織A及組織B組成且DNA混合物C之總甲基化密度為60%，則吾人可根據下式推論組織A及B對DNA混合物C之比例貢獻： MDC = MDA × α + MDB × b， 其中MDA 、MDB 、MDC 分別表示組織A、組織B及DNA混合物C之MD；且a及b為組織A及B對DNA混合物C之比例貢獻。在此特定實例中，假定組織A及B為DNA混合物僅有的兩種成分。因此，a+b=100%。因此據計算，組織A及B分別對DNA混合物貢獻60%及40%。 組織A及組織B中之甲基化密度可獲自生物體樣品或獲自相同類型之其他生物體(例如其他人類，潛在地為相同亞群)之樣品。若使用來自其他生物體之樣品，則組織A之樣品之甲基化密度之統計分析(例如平均值、中值、幾何平均值)可用於獲得甲基化密度MDA ，且對於MDB 情況類似。 可選擇基因組位點以具有最小個體間變化，例如小於變化之特定絕對量或在測試之基因組位點之最低部分內。舉例而言，對於最低部分，實施例可僅選擇測試之一群基因組位點中具有變化之最低10%之基因組位點。其他生物體可獲自健康個人，以及患有特定生理病況之個人(例如懷孕女性、或不同年齡的人或特定性別的人)，其可對應於包括測試的當前生物體之特定亞群。 亞群之其他生物體亦可具有其他病理學病況(例如患有肝炎或糖尿病等的患者)。此類亞群可對於各種組織具有改變的組織特異性甲基化模式。除使用正常組織之甲基化模式以外，此類疾病病況下之組織之甲基化模式可用於解卷積分析。此解卷積分析可在測試來自患有彼等病況之此類亞群之生物體時較精確。舉例而言，硬化肝或纖維化腎可具有分別相比於正常肝臟及正常腎臟之不同甲基化模式。因此，若對於肝硬化患者篩選其他疾病，則可較精確的是包括硬化肝作為向血漿DNA貢獻DNA之候選物中的一者，連同其他組織類型之健康組織。B . 多個基因組位點 當存在更多潛在候選組織時，更多基因組位點(例如10個或大於10個)可用於測定DNA混合物之構成。DNA混合物之比例組成之評估精確性取決於多種因素，包括基因組位點數目、基因組位點(亦稱作「位點」)對特定組織之特異性及跨越不同候選組織及跨越用於測定參考組織特異性程度之不同個體之位點變化性。位點對組織之特異性係指特定組織於其他組織類型之間的基因組位點之甲基化密度差異。 其甲基化密度之間的差異愈大，位點將對特定組織愈具特異性。舉例而言，若位點在肝臟中完全甲基化(甲基化密度=100%)且在所有其他組織中完全未甲基化(甲基化密度=0%)，則此位點將對肝臟具有高度特異性。然而，跨越不同組織之位點變化性可由例如(但不限於)不同類型的組織中之位點之甲基化密度之範圍或標準差反映。較大範圍或較高標準差將允許在數學上更確切及精確地測定不同器官對DNA混合物之相對貢獻。此等因素對估計候選組織對DNA混合物之比例貢獻之精確性的影響說明於本申請案之隨後部分中。 此處，吾人使用數學方程式以說明不同器官對DNA混合物之比例貢獻的推演。DNA混合物中之不同位點之甲基化密度與不同組織中之對應位點之甲基化密度之間的數學關係可表示為：

Figure 02_image005
, 其中
Figure 02_image007
表示DNA混合物中之位點i之甲基化密度；pk 表示組織k對DNA混合物之比例貢獻；MDik 表示組織k中之位點i之甲基化密度。當位點數目與器官數目相同或大於該數目時，可測定個別pk 之值。組織特異性甲基化密度可獲自其他個體，且可選擇位點以具有最小個體間變化，如上所述。 額外標準可包括於算法中以改良精確性。舉例而言，所有組織之聚集貢獻可受限於100%，即
Figure 02_image009
。 此外，所有器官之貢獻可需要為非負的：
Figure 02_image011
由於生物學變異，觀測之總體甲基化模式可能與推論自組織之甲基化之甲基化模式不完全相同。在此類情況下，將需要數學分析以測定個別組織之最可能的比例貢獻。就此而言，DNA中觀測之甲基化模式於自組織推論之甲基化模式之間的差值藉由W指示。
Figure 02_image013
其中O為對於DNA混合物觀測之甲基化模式且Mk 為個別組織k之甲基化模式。pk 為組織k對DNA混合物之比例貢獻。各pk 之最可能的值可藉由使W最小化測定，其為觀測與推論之甲基化模式之間的差值。此方程式可使用數學算法求解，例如藉由(但不限於)使用二次規劃、線性/非線性回歸、期望最大化(EM)算法、最大似然性算法、最大後驗評估及最小平方方法。C . 甲基化解卷積方法 如上文所述，可分析包括來自生物體之游離之DNA分子之混合物的生物樣品以測定混合物之組成，特定言之來自不同組織類型之貢獻。舉例而言，可測定來自肝臟之游離之DNA分子之百分比貢獻。生物樣品中之百分比貢獻之此等量測值可用於進行生物樣品之其他量測，例如鑑別腫瘤所位於之位置，如在稍後部分中描述。 圖1為說明分析游離之DNA分子之DNA混合物以自根據本發明之實施例之甲基化程度測定來自各種組織類型之百分比貢獻之方法100的流程圖。生物樣品包括來自M種組織類型之游離之DNA分子之混合物。生物樣品可為各種實例中的任一者，例如如本文中所提及。組織類型之數目M大於2。在各種實施例中，M可為3、7、10、20或大於20，或中間的任何數目。方法100可至少部分由使用電腦系統進行，本文所描述之其他方法亦可如此。 在步驟110處，對於分析鑑別N個基因組位點。N個基因組位點可具有各種屬性，例如如第II部分中更詳細地描述，該部分描述I型及II型基因組位點。作為實例，N個基因組位點可僅包括I型或II型位點，或兩者之組合。可基於一或多種其他樣品之分析，例如基於獲自關於各種個體中量測之甲基化程度之資料庫之資料鑑別基因組位點。 可選擇特定基因組位點以提供所需精確性程度。舉例而言，可使用具有至少一個臨限值變化性之基因座，與僅使用對一種組織類型具有特異性之基因座相反。可選擇第一組(例如10個)基因組位點以使得各自具有跨越M種組織類型至少0.15之甲基化程度之變異係數且使得各自具有就一或多種其他樣品而言超過0.1的M種組織類型之最大與最小甲基化程度之間的差異。此第一組基因組位點可不對於特定組織類型具有特定甲基化簽名，例如僅僅或主要在特定組織類型中經甲基化。此類第一組稱為II型位點。此等基因組位點可與稱為I型位點的具有特定簽名之基因組位點組合使用。 使用II型位點可確保跨越組織類型之甲基化程度之全空間由基因組位點橫跨，進而提供相比於I型位點增加之精確性。僅使用更多I型位點對甲基化空間提供冗餘基矢(亦即更多與其他位點具有相同模式之基因組位點)，而添加甲基化程度具有跨越不同組織之各種值之其他基因組位點添加新穎基矢以經由線性方程組區分百分比貢獻。 在一些實施例中，N個基因組位點中之至少10個各自具有至少0.15的跨越M種組織類型之甲基化程度之變異係數。至少10個基因組位點亦可各自具有超過0.1的M種組織類型之最大與最小甲基化程度之間的差異。可對於一個樣品或樣品組量測基因座之此等甲基化特性。樣品組可針對包括測試之本發明生物體之生物體亞群，例如具有與本發明生物體共用之特定特點之亞群。此等其他樣品可稱作參考組織，且可使用來自不同樣品之不同參考組織。 在步驟120處，在M中組織類型中之每一者之N個基因組位點獲得N個組織特異性甲基化程度。N大於或等於M，以使得組織特異性甲基化程度可用於解卷積以測定分率百分比。組織特異性甲基化程度可形成維度N×M之矩陣A。矩陣A之各行可對應於特定組織類型之甲基化模式，其中模式具有N個基因組位點之甲基化程度。 在各種實施例中，組織特異性甲基化模式可自公共資料庫或前述研究檢索。在本文中之實例中，嗜中性白血球及B細胞之甲基化資料自基因表現彙編(Gene Expression Omnibus)(Hodges等人 Mol Cell 2011;44:17-28)下載。其他組織(海馬區、肝、肺、胰臟、心房、結腸(包括其各種部分，例如乙狀結腸、橫結腸、升結腸、降結腸)、腎上腺、食道、小腸及CD4 T細胞)之甲基化模式自RoadMap Epigenomics項目(Ziller等人 Nature 2013; 500:477-81)下載。白血球層、胎盤、腫瘤及血漿資料之甲基化模式自報導公佈(Lun等人 Clin Chem. 2013;59:1583-94；Chan等人 Proc Natl Acad Sci USA. 2013;110:18761-8)。此等組織特異性甲基化模式可用於鑑別用於解卷積分析中之N個基因組位點。 在步驟130處，接收包括來自M種組織類型之游離之DNA分子之混合物的生物樣品。生物樣品可以多種方法獲自患者生物體。獲得此類樣品之方式可為非侵入性或侵入性的。非侵入性獲得之樣品之實例包括某些類型之流體(例如血漿或血清或尿液)或糞便。舉例而言，血漿包括來自許多器官組織之游離之DNA分子，且因此適用於經由一種樣品分析許多器官。 在步驟140處，來自生物樣品之游離之DNA分子經分析以鑑別其對應於生物體之參考基因組中之位置。舉例而言，游離之DNA分子可經測序以獲得序列讀段，且可將該等序列讀段與參考基因組映射(比對)。若生物體為人類，則參考基因體將為潛在地來自特定子群體之參考人類基因體。作為另一實例，可藉由不同探針分析游離之DNA分子(例如在PCR或其他擴增之後)，其中各探針對應於基因組位置，其可覆蓋雜合子及一或多個CpG位點，如下文所述。 可分析統計顯著數目之游離之DNA分子以提供精確解卷積以測定來自M種組織類型之百分比貢獻。在一些實施例中，分析至少1,000個游離之DNA分子。在其他實施例中，可分析至少10,000或50,000或100,000或500,000或1,000,000或5,000,000個或超過5,000,000個游離之DNA分子。分析之分子總數可取決於M及N，及所需精確度(精確性)。在各種實例中，分析之游離DNA之總數可小於500,000、1百萬、2百萬、5百萬、1千萬、2千萬或5千萬。 在步驟150處，使用各位於參考基因組之N個基因組位點中的任一者處之游離之DNA分子量測N個基因組位點之N混合物甲基化程度。DNA分子可藉由對應於基因組位點或基因座之一或多個鹼基位置之DNA分子之一或多個鹼基鑑別為位於基因組位點或基因座處。因此，DNA分子之序列將覆蓋基因組位點或基因座之一或多個鹼基位置。此資訊可基於步驟140中測定之位置測定。位於基因座位點之DNA分子之此類鑑別可用於本文所述之方法之任何類似步驟。 N混合物甲基化程度係指生物樣品之混合物中之甲基化程度。舉例而言，若來自混合物之游離之DNA分子位於N個基因組位點中的一者處，則該位點之分子之甲基化指數可包括於該位點之總甲基化密度中。N混合物甲基化程度可形成長度為N之甲基化向量b，其中b對應於可測定各對應組織類型之百分比貢獻之觀測值。 在一個實施例中，可使用全基因組亞硫酸氫鹽定序測定DNA混合物中之基因組位點之甲基化程度。在其他實施例中，可使用甲基化微陣列分析，如Illumina HumanMethylation450系統，或藉由使用甲基化免疫沈澱(例如使用抗甲基胞嘧啶抗體)或用甲基化結合蛋白處理，繼之以微陣列分析或DNA定序，或藉由使用甲基化敏感限制酶處理，繼之以微陣列或DNA定序，或藉由使用甲基化感測定序，例如使用單分子定序方法(例如藉由奈米孔定序(Schreiber等人 Proc Natl Acad Sci 2013; 110: 18910-18915)或藉由Pacific Biosciences單分子實時分析(Flusberg等人 Nat Methods 2010; 7: 461-465))測定CpG位點之甲基化程度。組織特異性甲基化程度可以相同方式量測。在其他實施例中，其他方法，例如(但不限於)靶向亞硫酸氫鹽定序、甲基化特異性PCR、基於非亞硫酸氫鹽之甲基化感測定序(例如藉由單分子定序平台(Powers等人 Efficient and accurate whole genome assembly and methylome profiling of E. coli. BMC Genomics. 2013; 14:675))可用於分析血漿DNA甲基化解卷積分析之血漿DNA之甲基化程度。 在步驟160處，測定組合向量之M值。各M值對應於M中組織類型之特定組織類型對DNA混合物之百分比貢獻。鑒於矩陣A由N×M組織特異性甲基化程度組成(亦即M種組織類型中之每一者之N個組織特異性甲基化程度)，可求解組合向量之M值以提供N混合物甲基化程度(例如甲基化向量b)。M百分比貢獻可對應於藉由求解Ax=b確定之向量x。當N大於M時，求解可涉及誤差之最小化，例如使用最小平方。 在步驟170處，使用組合向量測定混合物中之M種組織類型中之每一者之量。組合向量之M值可直接當作M種組織類型之百分比貢獻。在一些實施方案中，M值可轉化為百分比。誤差術語可用於將M值轉換至較高或較低值。D . 應用 如上所述，百分比貢獻可用於生物樣品之其他量測及其他測定，例如特定染色體區域是否具有序列不平衡、特定組織類型是否患病及測定由獲得樣品之懷孕女性之胎兒遺傳兩種親體單倍型中之何種單倍型。 圖2顯示顯示根據本發明之實施例之DNA甲基化解卷積(例如使用血漿)之若干潛在應用之示意圖。圖2中，生物樣品205在210處進行全基因組亞硫酸氫鹽定序。在230處，血漿DNA組織映射使用組織特異性甲基化圖譜220測定組織貢獻百分比。實例組織特異性甲基化圖譜顯示為肝臟、血細胞、脂肪組織、肺、小腸及結腸。貢獻百分比可如上文及他處所述測定，例如求解Ax=b。應用之實例包括產前測試241、癌症偵測及監測242、器官移植監測及器官損傷評估244。 適用於測定不同器官對血漿DNA之貢獻的甲基化標記(基因組位點)之清單可藉由比較不同組織，包括肝臟、肺、食道、心臟、胰臟、乙狀結腸、小腸、脂肪組織、腎上腺、結腸、T細胞、B細胞、嗜中性白血球、大腦及胎盤之甲基化特徵(圖2)鑑別。在各種實例中，肝臟、肺、食道、心臟、胰臟、結腸、小腸、脂肪組織、腎上腺、大腦及T細胞之全基因組亞硫酸氫鹽定序資料自來自拜勒醫藥學會(Baylor College of Medicine)之人類表觀基因組圖譜(Human Epigenome Atlas)檢索(www.genboree.org/epigenomeatlas/index.rhtml)。B細胞及嗜中性白血球之亞硫酸氫鹽定序資料來自Hodges等人之出版物(Hodges等人; Directional DNA methylation changes and complex intermediate states accompany lineage specificity in the adult hematopoietic compartment. Mol Cell 2011; 44: 17-28)。胎盤之亞硫酸氫鹽定序資料來自Lun等人(Lun等人 Clin Chem 2013; 59:1583-94)。在其他實施例中，標記物可鑑別自使用微陣列分析，例如使用Illumina Infinium HumanMethylation450 BeadChip陣列產生之資料集。 II.甲基化標記物之選擇 在上文中，吾人已描述使用甲基化分析測定DNA混合物之組成的原理。特定言之，不同器官(或組織)對血漿DNA之百分比貢獻可使用甲基化分析測定。在此部分中，吾人另外描述選擇甲基化標記物之方法及此技術之臨床應用。 藉由甲基化分析測定DNA混合物之組成的結果受用於DNA混合物之組成之解卷積的甲基化標記物影響。因此，選擇適當基因組甲基化標記物可就精確測定DNA混合物之構成而言重要。A . 用於解卷積之甲基化標記物之標準 對於標記物選擇，可考慮以下三中屬性：(i)甲基化標記物需要具有在跨越不同個體之相同組織類型中量測之甲基化程度之低變化性。由於DNA混合物之組成的測定取決於組織特異性甲基化模式之識別，跨越不同個體之相同組織類型中之甲基化程度之低變化性將適用於精確鑑別DNA混合物中之組織特異性模式。在組織特異性甲基化程度獲自其他生物體之樣品(例如獲自資料庫)之實施例中，低變化性意謂來自其他樣品之甲基化程度與當前測試之生物體之組織特異性甲基化程度類似。 (ii)甲基化標記物需要具有跨越不同組織之甲基化程度之高變化性。對於特定標記物，跨越不同組織之甲基化程度之較高差異可提供不同組織對DNA混合物之貢獻的更精確測定。特定言之，可藉由使用具有屬性(ii)之一族標記物及具有屬性(iii)之另一組標記物獲得精確性之改進。 (iii)甲基化標記物需要具有相比於大部分或所有其他組織時特定不同的特定組織中之甲基化程度。相比於上文第(ii)點，標記物可具有大部分組織之甲基化程度之低變化性，但其在一種特定組織中之甲基化程度不同於大部分其他組織。此標記物將特別適用於測定與其他組織具有不同甲基化程度之組織的貢獻。B . 實例 標記物選擇之原理說明於表1中之以下假設實例中。    標記物1 標記物2 標記物3 標記物4 標記物5 標記物6 肝臟1 20% 69% 9% 9% 10% 90% 肝臟2 50% 70% 10% 10% 10% 90% 肝臟3 90% 71% 11% 11% 10% 90% 心臟 20% 20% 30% 13% 12% 12% 肺 30% 30% 60% 17% 14% 84% 結腸 40% 40% 90% 20% 80% 80% 表 1 . 6種假設甲基化標記物之不同組織中之甲基化密度。 在此假設實例中，當相比於標記物1時，標記物2在來自三個個體之肝臟中之甲基化密度中具有較低變化性。因此，作為測定肝臟在DNA混合物中之貢獻的簽名，標記物2優於標記物1。 相比於標記物4，標記物3具有跨越不同組織類型之甲基化密度之較高變化性。根據上文所述之數學關係，來自不同組織之估計貢獻之相同變化程度將對於標記物3提供相比於標記物4更大的DNA混合物之推論甲基化密度之變化。因此，各組織之貢獻之評估可在標記物3之情況下更精確。 標記物5具有跨越肝臟、心臟及肺之甲基化密度之低變化性。其甲基化密度在10%至14%之範圍內變化。然而，結腸之甲基化密度為80%。此標記物將特別適用於測定結腸在DNA混合物中之貢獻。類似地，相比於針對標記物6之其他組織，心臟為低甲基化的。因此，心臟之貢獻可藉由標記物6精確測定。因此，標記物5及6之組合將能夠精確測定結腸及心臟之貢獻。添加標記物2及3將接著足以推論包括肝臟、心臟、肺及結腸之四種器官中之每一者之貢獻。C . 不同類型的標記物 甲基化標記物可不必需具有所有以上三種屬性。I型甲基化標記物將通常具有以上屬性(iii)。多種此類標記物亦可具有屬性(i)。另一方面，II型甲基化標記物將通常具有以上屬性(ii)。多種此類標記物亦可具有屬性(i)。亦有可能的是特定標記物可具有全部三種屬性。 在一些實施例中，標記物廣泛地分成兩種類型(I型及II型)。I型標記物具有組織特異性。此等標記物對特定群組之一或多種組織之甲基化程度不同於大部分其他組織。舉例而言，特定組織可具有相比於所有其他組織之甲基化程度的顯著甲基化程度。在另一實例中，兩種組織(例如組織A及組織B)具有類似甲基化程度，但組織A及B之甲基化程度顯著不同於其餘組織之甲基化程度。 II型標記物具有高組織間甲基化變化性。此等標記物之甲基化程度跨越不同組織高度可變。此類別中之單一標記物可能不足以測定特定組織對DNA混合物之貢獻。然而，II型標記物之組合，或與一或多種I型標記物之組合可共同地使用以推論個別組織之貢獻。在以上定義下，特定標記物可僅為I型標記物、僅為II型標記物或同時為I型及II型標記物兩者。1. I 型 標記物 在一個實施例中，可藉由對於所有候選組織比較標記物之甲基化密度與此特定標記物之甲基化密度之平均值及標準差(SD)鑑別I型標記物。在一個實施方案中，若標記物在一個組織中之甲基化密度不同於所有組織之平均值3個標準差(SD)，則標記物經鑑別。 研究獲自上文所提及之來源之14種組織之甲基化圖譜以選擇標記物。在一個分析中，使用以上標準鑑別總共1,013種I型標記物(美國臨時申請案第62/158,466號之附錄A之表S1中標記為I型之標記物)。在其他實施例中，可使用特定組織與平均甲基化密度之間的其他閾值，例如(但不限於)1.5 SD、2 SD、2.5 SD、3.5 SD及4 SD。在另一實施例中，I型標記物可經由特定組織之甲基化密度與所有組織之中值甲基化密度之比較鑑別。 在其他實施例中，當超過一個組織(例如(但不限於)兩個、三個、四個或五個組織)顯示與所有候選組織之平均甲基化密度顯著不同的甲基化密度時，可獲得I型標記物。在一個實施方案中，截止甲基化密度可計算自所有候選組織之甲基化密度的平均值及SD。出於說明目的，閾值可定義為比平均甲基化密度高或低3 SD。當超過一個(例如(但不限於)兩個、三個、四個、五個或大於五個)組織之甲基化密度高於組織之平均甲基化密度超過3 SD或低於平均甲基化密度超過3 SD時，選擇標記物。 2. II型標記物 為了鑑別II型標記物，計算跨越所有14種候選組織之甲基化密度之平均值及SD且SD與平均值的比率表示為變異係數(CV)。在此說明性實例中，吾人對於CV使用＞0.25之閾值以鑑別合格II型標記物，以及超出0.2之組織群之最大與最小甲基化密度之間的差值。使用此等標準，鑑別5820種II型標記物(附錄A之表S1中標記為II型之標記物)。在其他實施例中，可對於CV使用其他閾值，例如(但不限於)0.15、0.2、0.3及0.4。在其他實施例中，可使用最大與最小甲基化密度之間的差值之其他閾值，例如(但不限於)0.1、0.15、0.25、0.3、0.35、0.4、0.45及0.5。 在其他實施例中跨越相同組織類型之多個樣品之平均值可用於量測跨越不同組織之甲基化程度之變化。舉例而言，可對來自10個樣品之相同基因組位點之10個甲基化程度取平均值以獲得基因組位點之單一甲基化程度。可進行類似方法以對於基因組位點之其他組織類型測定平均甲基化程度。跨越組織類型之平均值可隨後用於測定基因組位點是否具有跨越組織類型之顯著變化。可使用除平均值以外的其他統計值，例如中值或幾何平均值。此類統計值可用於鑑別I型及/或II型標記物。 相同組織類型之不同樣品(例如來自不同個體)可用於測定跨越不同樣品之甲基化程度之變化。因此，若存在多個相同組織類型之樣品，則實施例可另外量測相同組織類型之此類樣品之間的特定標記物之變化。具有跨越樣品之低變化之標記物將為比具有高變化之標記物更可靠的標記物。標記物及解卷積之其他細節可見於Chiu等人之名稱為「Methylation Pattern Analysis Of Tissues In A DNA Mixture」之共同擁有的美國專利公開案2016/0017419及名稱為「Non-Invasive Determination Of Methylome Of Fetus Or Tumor From Plasma」之PCT公開案WO2014/043763中。D . 不同類別之標記物 基因組基因座(甲基化標記物)之「類別」對應於跨越相同組織類型之不同個體之基因座之甲基化程度之特定變化。不同類別可在跨越個體之特定組織類型之間具有不同變化範圍。第一類別之甲基化標記物可在測試之個體之間具有10%或更低的甲基化程度之差異。第二類別之甲基化標記物可在測試之個體之間具有大於10%之甲基化程度之差異。使用具有低個體間變化之甲基化標記物(第一類別標記物)將潛在地改良測定特定器官在DNA混合物中之貢獻之精確性。E . 鑑別潛在甲基化標記物 在一些實施例中，以如下方式鑑別潛在甲基化標記物。此類潛在甲基化標記物可隨後經受以上標準以鑑別I型及II型標記物。在其他實施例中，不需要鑑別I型或II型。且其他實施例可使用其他技術鑑別潛在甲基化標記物。 在一些實施例中，對於潛在甲基化標記物考慮常染色體上之CpG島(CGI)及CpG岸。不使用性染色體上之CGI及CpG岸以使與源資料中之性別相關染色體劑量差異有關的甲基化程度之變化最小化。CGI自加州大學聖塔克魯斯(UCSC)資料庫(genome.ucsc.edu/, 27,048個關於人類基因組之CpG島)下載(Kent等人, UCSC之人類基因組瀏覽器, Genome Res. 2002;12(6):996-1006)且CpG岸定義為CpG島之2 kb邊窗(Irizarry等人 The human colon cancer methylome shows similar hypo- and hypermethylation at conserved tissue- specific CpG island shores. Nat Genet 2009; 41(2): 178 - 186)。接著，CpG島及岸細分為非重疊500 bp單元且各單元視為潛在甲基化標記物。 在14種組織類型之間比較所有潛在基因座之甲基化密度(亦即在500 bp單元內甲基化CpGs之百分比)。如先前報導(Lun等人Clin Chem. 2013; 59: 1583-94)，發現胎盤與其餘組織比較時整體低甲基化。因此，胎盤之甲基化圖譜不包括於標記物鑑別階段。使用其餘13種組織類型之甲基化圖譜，鑑別兩種類型之甲基化標記物。舉例而言，當與13種組織類型之平均值比較時，I型標記物可指在一種組織中具有低於或高於3 SD之甲基化密度的任何基因組位點。當(A)最高甲基化組織之甲基化密度高於最低甲基化組織之甲基化密度至少20%；及(B)跨越13種組織類型之甲基化密度除以該群之平均甲基化密度之SD (亦即變異係數)為至少0.25時，II型標記物可視為高度可變。最後，為了減少潛在冗餘標記物之數目，在兩個CpG岸側接一個CpG島之一個連續區段中僅可選擇一個標記物。F . 基於應用之選擇 對於特定應用選擇之甲基化標記物集合可依所需應用之參數變化。舉例而言，對於聚焦於單倍型或等位基因分析之應用，適用標記物將為該等位於相同游離之DNA分子上為雜合等位基因中之一者。由於游離之DNA分子(例如血漿DNA)通常小於200 bp，適用標記物可為200 bp之雜合基因座(例如SNP)內之CpG位點。作為另一實例，對於DNA自特定組織釋入血漿具有特殊重要性之應用，吾人可選擇優先較大數目之甲基化標記物，當與標記物集合中之其他標記物比較時，在此組織類型中差別地甲基化(例如I型標記物)。 解卷積分析中之甲基化標記物之數目及選擇可根據預期用途變化。若對肝臟之百分比貢獻特別感興趣，例如在已接受肝臟移植之患者中，則更多I型肝臟特異性標記物可用於解卷積分析以增加移植肝臟對血漿DNA之貢獻之定量的精確性。 III. 組成精確性 如上文所述，實施例可鑑別血漿DNA之組織貢獻者。在各種實例中，進行血漿DNA之全基因組亞硫酸氫鹽定序且參看不同組織之甲基化圖譜進行分析。使用二次規劃作為實例，血漿DNA定序資料去卷積為來自不同組織之比例貢獻。對於懷孕女性；患有肝細胞癌、肺癌及結腸直腸癌之患者；以及骨髓及肝臟移植後的個體測試實施例。 在大部分個體中，白血球為循環DNA池之主要貢獻者。懷孕女性中之胎盤貢獻與如藉由胎兒特異性遺傳標記物顯示之比例貢獻相關。移植衍生的對移植受體中之血漿之貢獻與使用供體特異性遺傳標記物測定之彼等相關。患有肝細胞癌、肺癌或結腸直腸癌之患者顯示來自具有腫瘤之器官之較高血漿DNA貢獻。肝細胞癌患者中之肝臟貢獻亦與使用腫瘤相關拷貝數變異進行之量測相關。 在癌症患者及展現血漿中之拷貝數變異之懷孕女性中，甲基化解卷積識別造成偏差之組織類型。在懷孕期間診斷為患有濾泡性淋巴瘤之懷孕女性中，甲基化解卷積指示大體上較高的自B細胞向血漿DNA池之貢獻且將局部B細胞(而非胎盤)指示為血漿中觀測之拷貝數變異之來源。因此，實施例可充當基於不同組織對血漿之擾動比例貢獻之鑑別評估大範圍的生理學及病理學病況之強力工具。A . 不同類型的血球之貢獻 作為甲基化解卷積之實例，吾人測定不同組織及細胞類型對循環DNA之貢獻。自兩個罹患全身性紅斑性狼瘡症(SLE)之患者收集兩個血液樣品。在收集之後，靜脈血液樣品在1,500 g下離心10分鐘。在離心之後，分離血細胞及血漿。接著自血細胞提取DNA。DNA經亞硫酸氫鹽轉化且使用HiSeq2000定序器中之流動池之一條通道定序。使用細胞類型特異性甲基化模式分析來分析兩個血細胞樣品。嗜中性白血球、淋巴球、食道、結腸、胰臟、肝臟、肺、心臟、腎上腺及海馬區之甲基化模式包括為血細胞DNA之潛在候選物。609 選擇甲基化標記物用於分析。亦將兩個個體之全血樣品送至細胞計數以測定血細胞之嗜中性白血球及淋巴球之分率組成。    血液樣品1 血液樣品2    細胞類型特異性甲基化模式分析 血球計數 細胞類型特異性甲基化模式分析 血球計數 嗜中性白血球 90.5% 93.6% 89.4% 89.9% 淋巴球 9.5% 6.4% 10.6% 10.1% 食道 0% - 0% - 結腸 0% - 2% - 胰臟 0% - 0% - 肝臟 0% - 1% - 肺 1% - 1% - 心臟 0% - 3% - 腎上腺 0% - 0% - 海馬區 0% - 0% - 表 2 . 藉由解卷積模式分析及細胞計數之血液組織貢獻 對於甲基化模式分析，嗜中性白血球及淋巴球測定為構成血細胞DNA之主要組分。嗜中性白血球及淋巴球之貢獻之相對比例與其根據細胞計數分析之血液樣品中之相對豐度類似。B . 懷孕女性 使用懷孕女性之血漿DNA之甲基化分析來分析不同組織，包括肝臟、肺、胰臟、結腸、海馬區、小腸、血細胞、心臟、腎上腺、食道及胎盤之貢獻。由於胎盤基因型一般與胎兒的基因型相同但不同於懷孕女性的基因型，胎盤對母體血漿之精確貢獻可藉由計數樣品照組胎兒特異性等位基因數目精確測定。1. 組成及與胎兒 DNA 百分比之相關性 對於15個懷孕女性(來自第一、第二及第三個三月期中之每一者之五個)進行血漿DNA之全基因組亞硫酸氫鹽定序。進行甲基化解卷積且推論來自不同組織之百分比貢獻。基於使用二次規劃分析之表S1中之所有I型及II型標記物之甲基化程度(如甲基化密度)測定不同器官之貢獻。 圖3A顯示根據本發明之實施例之就15個懷孕女性而言之不同器官對血漿DNA之百分比貢獻的圖300。各條形對應於一種樣品之結果。不同顏色表示不同器官對血漿之貢獻。此等結果顯示白血球(亦即嗜中性白血球及淋巴球)為血漿DNA池之最重要貢獻者。此觀測結果與先前在骨髓移植後獲得之觀測結果一致((Lui YY等人 Clin Chem 2002; 48:421-7)。 圖4顯示根據本發明之實施例測定自懷孕女性中之血漿DNA組織映射分析之百分比貢獻的表400。此等結果亦顯示胎盤為懷孕女性中之血漿DNA之另一關鍵貢獻者，分率濃度為9.9%至38.4%。 吾人亦使用如先前所述之懷孕女性不具有之父本遺傳胎兒單核苷酸多態性(SNP)等位基因量測胎盤貢獻(31)。為了分析胎兒特異性SNP等位基因，藉由分析絨毛膜絨毛樣品或胎盤測定胎兒之基因型。藉由分析血細胞測定懷孕女性之基因型。基於SNP之結果顯示甲基化解卷積結果之獨立驗證。 圖3B顯示推論自血漿DNA甲基化解卷積之胎盤貢獻之血漿DNA分率與使用根據本發明之實施例的胎兒特異性SNP等位基因推論之胎兒DNA百分比濃度之間的相關性之圖350。圖350顯示藉由甲基化解卷積測定之胎盤貢獻與使用SNP量測之胎兒DNA百分比濃度具有強相關性(r =0.99，p＜0.001，皮爾森相關性(Pearson correlation))。因此，在兩個參數之值之間觀測到良好正相關，表明血漿DNA甲基化解卷積精確測定胎盤對母體血漿樣品之貢獻。 圖5顯示根據血漿DNA組織映射的除胎盤以外之器官之百分比貢獻及基於根據本發明之實施例的胎兒特異性SNP等位基因之胎兒DNA百分比濃度之圖。X軸表示藉由基於SNP之分析估計之胎兒DNA百分比濃度且Y軸表示藉由血漿組織DNA映射分析推論之百分比貢獻。嗜中性白血球之血漿DNA貢獻顯示逆相關。此可能由於嗜中性白血球為血漿DNA池之主要貢獻者且因此胎盤貢獻增加，來自嗜中性白血球之相對貢獻將必然地減少。其餘組織之甲基化解卷積結果不顯示與胎兒DNA百分比濃度之相關性。 圖6顯示來自根據本發明之實施例之非懷孕健康對照個體組之血漿DNA組織映射分析之百分比貢獻的表600。當方法應用於非懷孕健康對照之血漿時，在大部分樣品中不存在胎盤貢獻(中值：0%；四分位數範圍：0%至0.3%)。2. 所選標記物與隨機標記物之比較 藉由所選標記物相對於隨機標記物測試百分比貢獻之精確性。對於不同標記物組進行不同組成計算。基於上文提及之標準選擇一組，且另一組為隨機組。結果顯示重要的是公正地選擇甲基化標記物(基因座)用途，以便獲得精確結果。 對於此分析募集十一個懷孕女性及四個健康非懷孕個體。其血漿DNA經亞硫酸氫鹽轉化且使用Illumina HiSeq2000定序器定序。各血漿樣品藉由定序流動池之一個通道定序。接著使用生物信息學程式Methy-Pipe分析序列讀數(Jiang P. PLoS One 2014; 9: el00360)。此程式可將亞硫酸氫鹽轉化序列讀數與參考基因組比對且測定各定序片段上之各CpG位點之甲基化狀態。 第一組標記物對於鑑別血漿DNA中之不同組織具有高特異性。對於各組織類型，選擇相比於其他組織具有甲基化密度之最大差異的標記物。標記物測定自含有至少一個CpG二核苷酸之基因組區域。在此實例中，CpG島(CGI)用作潛在標記物，其在DNA之特定伸長部中具有高頻之CpG位點。此特定實例中之CGI下載自加州大學聖塔克魯斯(UCSC)資料庫：(genome.ucsc.edu)。總計，吾人自人類基因組獲得27,048個CpG島。CpG島之中值尺寸為565 bp(範圍：200 bp至45 kb)。90%之島小於1.5 kb。 對於各甲基化標記物，測定所關注的組織與其他組織之間的甲基化密度之差值。差值接著表示為跨越其他組織之標準差(SD)數目。對於所關注的組織，所有標記物根據甲基化密度之此差值分級。選擇具有高於(10個標記物)及低於(10個標記物)其他組織之平均甲基化密度之最大差異的20個標記物。標記物之數目可變化，例如(但不限於)5、15、20、30、40、50、100及200。 另外，亦選擇具有跨越所有不同組織之高變化性的標記物。在此實例中，選擇在具有最高與最低甲基化密度之組織之間具有＞50%差值之標記物。在其他應用中，可使用其他值，例如(但不限於)20%、30%、40%、60%、70%及80%。此外，亦基於平均值及SD計算跨越不同組織之甲基化密度之變化性。在此實例中，若SD之值超過平均值兩倍，則亦選擇標記物。在其他應用中，亦可使用其他閾值，例如(但不限於)1、1.5、2.5及3。基於此等選擇標準，對於第一組選擇344個甲基化標記物。 對於第二組，341個標記物隨機選自上文所述之27,048個CGI。所有CGI首先編號為1至27,048。接著，由電腦產生隨機數(在1與27,048之間)用於標記物選擇。接著重複此方法直至選擇總共341個標記物。若產生之隨機數已使用，則將產生另一隨機數。預期此組標記物在鑑別組織特異性甲基化模式中具有低得多的特異性。因此，預期測定血漿DNA之組成的精確性降低。 圖7顯示關於根據本發明之實施例使用第一組標記物(具有高器官特異性)之11個懷孕女性及4個非懷孕健康個體的不同器官對血漿DNA之估計貢獻之表700。藉由計數胎兒特異性等位基因測定胎兒DNA百分比濃度且顯示於底列中。在四個非懷孕對照個體中之每一者中，胎盤對血漿之貢獻測定為接近0%。此指示此方法之特異性。 圖8顯示關於根據本發明之實施例使用第二組標記物(具有低器官特異性)之11個懷孕女性及4個非懷孕健康個體的不同器官對血漿DNA之估計貢獻之表800。藉由計數胎兒特異性等位基因測定之胎兒DNA百分比濃度顯示於底列中。使用此等較低特異性標記物，觀測到自胎盤之相對非一致貢獻百分比，且在四個非懷孕對照個體中觀測到自胎盤之相當大的貢獻。此指示標記物之組織特異性在此方法中重要。 圖9A為顯示估計胎兒DNA百分比濃度(貢獻自胎盤)與藉由計數母體血漿樣品中之胎兒特異性等位基因測定之胎兒DNA百分比濃度之間的相關性之圖900。使用第一組甲基化標記物，來自兩種技術之結果具有良好相關性。然而使用第二組甲基化標記物，藉由使用甲基化分析之評估顯示與使用胎兒特異性等位基因計數測定之真實值之顯著偏差。 圖9B為顯示來自甲基化標記物之評估與藉由胎兒特異性等位基因計數測定之胎兒DNA百分比濃度之間的絕對差之圖950。使用甲基化分析之評估之中值誤差分別為使用第一組標記物及第二組標記物之4%及8%。C . 不同標準之影響 如上文所述，各種標準可用於鑑別不同類型之標記物。舉例而言，I型標記物可藉由不同於所有組織之平均甲基化程度(例如至少特定臨限值，如3 SD)之特定組織中之甲基化程度鑑別。且對於II型標記物，使用某一變化及最大差值之標準。以下部分顯示不同標準鑑別標記物之精確性。 1. 在較不嚴格標準下之標記物效能 吾人使用具有跨越不同組織之不同變化性的標記物比較甲基化解卷積分析之效能。基於具有不同選擇標準之兩組標記物對於15個懷孕女性測定胎盤對血漿DNA之貢獻。兩組標記物均包括如前述部分中所述之所有I型標記物。然而，兩組標記物之II型標記物之選擇標準不同。 第I組標記物包括所有5,820種滿足具有＞0.25之甲基化密度CV且組織群之最大與最小甲基化密度之間的差值超過0.2之標準的II型標記物。對於第II組標記物，CV要求為＞0.15且組織群之最大與最小甲基化密度之間的差值超過0.1。此組標記物中存在8,511種II型標記物。 圖10A為顯示根據本發明之實施例使用具有不同選擇標準之標記物推論的胎盤對血漿DNA之貢獻之圖1000。垂直軸對應於使用第II組標記物推論之胎盤貢獻。水平軸對應於使用第I組標記物推論之胎盤貢獻。在基於具有不同選擇標準之兩組標記物的胎盤貢獻結果之間存在良好相關性(r=0.99，皮爾森相關性)。因此，可使用CV＞0.15及組織群之最大與最小甲基化密度之間的差值超過0.1之要求獲得良好精確性。 2. 相同組織類型內之甲基化程度變化之影響 為了調查相同組織類型之間的標記物(例如來自不同個體)之甲基化程度變化是否將影響解卷積分析之效能，吾人分析來自兩個懷孕個案之胎盤組織。鑑別兩種類別之甲基化標記物。特定言之，兩種類別基於其在兩個胎盤組織中之甲基化程度之相似性而鑑別。類別i之標記物具有10%或低於10%之甲基化密度。類別ii之標記物在兩個胎盤組織之間具有高變化性(甲基化密度之差值大於10%)。 圖10B為顯示使用在相同組織類型中具有低變化性(類別i)及高變化性(類別ii)之標記物之血漿DNA解卷積之精確性的圖1050。進行血漿DNA解卷積以對於15個懷孕女性測定胎盤對血漿DNA之貢獻。對於各標記物，兩個胎盤組織之甲基化密度之平均值用於表示分析中之胎盤之甲基化程度。對於使用類別i及類別ii標記物之解卷積分析中之每一者，使用總共1,024種標記物。 另外基於胎兒特異性SNP等位基因之比例測定血漿中胎盤衍生之DNA的量。藉由基於類別i及類別ii標記物之甲基化解卷積分析推論之百分比貢獻接著相比於基於胎兒特異性SNP等位基因之結果。來自基於胎兒特異性等位基因估計之值的衍生胎盤貢獻之中值偏差分別為2.7%(使用類別i標記物)及7.1%(使用類別ii標記物)。因此，使用具有組織甲基化程度之較低個體間變化之類別i標記物給出甲基化解卷積分析中之較好精確性。 當使用在相同組織類型內具有高變化性之標記物(類別ii)時，觀測到來自甲基化解卷積與胎兒特異性等位基因分析之結果之間的顯著較高差異(P＜0.0001，威爾科克森符號秩檢驗(Wilcoxon sign-rank test))。換言之，使用在相同組織類型內具有低變化性之標記物將增加甲基化解卷積分析之精確性。因此，可基於相同組織類型內之變化性，例如(但不限於)CV值及相同組織類型之最大與最小甲基化密度之間的差值選擇標記物。IV. 胎兒簽名之解卷積 若已知基因組簽名(例如特定SNP等位基因)，則實施例可測定何種組織為此類簽名之來源。因此，若特定簽名代表胎兒(例如特定基因座之父系等位基因)，則簽名對胎盤組織之百分比貢獻將相當大。 為了說明單一核苷酸改變亦可用於測定衍生出改變之源組織，吾人分析懷孕女性之血漿DNA。胎盤及母體白血球層經基因分型以鑑別母親經純合且胎兒經雜合之SNP。吾人將胎兒及母親共用之等位基因指示為A且將胎兒特異性等位基因指示為B。因此，母親具有AA之基因型且胎兒在此等SNP中之每一者處具有AB之基因型。 在母體血漿DNA之亞硫酸氫鹽定序之後，選擇攜有胎兒特異性等位基因(B等位基因)及至少一個CpG位點之所有DNA片段且用於下游分析。總共13.1億片段經定序且677,140個攜有胎兒特異性等位基因(B等位基因)之片段用於解卷積分析。所有藉由至少10個DNA片段覆蓋之CpG位點用於解卷積分析。可使用覆蓋位點之其他數目的DNA片段，如5、15、20、25或30。由於B等位基因為胎兒特異性的，預期此等DNA片段衍生自胎盤。 組織 貢獻 ( % ) 肝臟 0.9 肺 0.0 結腸 0.0 小腸 0.0 胰臟 0.5 腎上腺 0.0 食道 3.1 脂肪組織 0.0 心臟 0.0 大腦 0.3 T細胞 0.0   B細胞 0.0   中性粒細胞 0.0   胎盤 95.2   表 3 . 使用胎兒特異性等位基因之甲基化解卷積分析。 在表3中，自甲基化解卷積分析顯示胎盤推論為攜有胎兒特異性SNP等位基因之此等DNA片段之主要貢獻者。此等結果表明甲基化解卷積分析精確鑑別攜有胎兒特異性等位基因之此等DNA片段之組織來源。 此顯示特定等位基因可歸於胎兒。此類技術更詳細地描述於下文中，用於使用甲基化解卷積分析測定胎兒之基因型及單倍型。V . 測定胎兒基因組 ( 突變分析 ) 對於非侵襲性產前測試，使用母體血漿DNA分析母體突變之遺傳為具挑戰性之任務。舉例而言，若懷孕女性對於突變為雜合的，則使用母體血漿DNA分析之胎兒突變狀態之分析將在技術上困難，因為不管其胎兒之突變狀態，突變及正常等位基因均將存在於其血漿中。先前，已開發多種不同方法以解決此問題(Lun等人 Proc Natl Acad Sci USA. 2008;105:19920-5；Lo 等人 Sci Transl Med. 2010;2:61ra91；Lam等人 Clin Chem. 2012;58:1467-75)。此等先前方法之原理涉及母體血漿中之突變與正常等位基因之相對量之間的比較。為了增強比較之統計功效，此等方法中的一些另外涉及連接至突變等位基因之SNP等位基因及連接至正常等位基因之彼等等位基因之相對量的比較。作為替代方案或另外，本發明之一些實施例可藉由甲基化解卷積分析推論胎兒之突變狀態。A . 使用甲基化解卷積之等位基因之貢獻 在此實例中，測定胎兒之基因型。假定父親及母親在特定基因座處之基因型分別為NN及MN。M及N分別指示突變及正常等位基因。在此情境下，胎兒可自母親遺傳M等位基因或N等位基因。因此，胎兒存在兩種可能的基因型，即MN及NN。在母體血漿中，攜有胎兒基因型之DNA實際上衍生自胎盤。因此，此等DNA片段將展現胎盤甲基化圖譜。 圖11A顯示胎兒遺傳來自母親之M等位基因且在根據本發明之實施例之特定基因座處具有MN之基因型的第一情形。在圖11A之頂部部分(經標記基因型)中，父親顯示為具有基因型NN，母親顯示為具有基因型MN，且胎兒顯示為具有基因型MN。展現胎盤甲基化圖譜之DNA片段用P標記，其顯示於胎兒基因型上。舉例而言，胎盤甲基化圖譜可對應於接近特定基因座之基因組位點之某些甲基化程度。對準至特定基因座之DNA片段亦可包括接近基因座之基因組位點(例如在基因座之200 bp內)，且因此可用於對於甲基化解卷積分析量測甲基化程度。考慮親體之基因型，M等位基因對母親具有特異性且N等位基因在父親與母親之間共用。 在圖11A之底部部分(經標記母體血漿)中，顯示兩個等位基因M及N之實例，其中各實例表示所關注的基因座處之血漿中之不同DNA分子。出於說明的目的，僅顯示少數DNA分子。在此實例中，胎兒DNA百分比假設為25%，如藉由25%之DNA分子用P標記所示。 在母體血漿樣品中，吾人選擇性地分析攜有M等位基因之DNA片段且進行甲基化解卷積分析。由於胎兒具有MN之基因型，胎盤將M及N等位基因兩者貢獻於母體血漿DNA。因此，攜有M等位基因之DNA片段中的一些亦將在接近基因座之基因組位點攜有胎盤特異性甲基化圖譜。甲基化解卷積分析將指示攜有M等位基因之DNA片段中的一些將衍生自胎盤，且因此胎兒基因型包括M等位基因。 圖11B顯示胎兒遺傳來自母親之N等位基因且在根據本發明之實施例之特定基因座處具有NN之基因型的第二情形。在此情況下，僅攜有N等位基因之DNA片段將在母體血漿中展現胎盤甲基化圖譜。因此，藉由甲基化解卷積之攜有M等位基因之DNA片段之選擇性分析將指示此等DNA片段不具有來自胎盤之顯著貢獻。因此，可確定胎兒不具有M，且因此具有NN之基因型。 在一些實施例中，可比較M及N等位基因之胎盤貢獻。此處，吾人假定胎兒DNA占總母體血漿DNA之大致10%。M及N等位基因之選擇性解卷積將適用於指示胎兒自母親遺傳何種等位基因。預期結果展示於下表4中：    胎兒基因型    MN NN 攜有M等位基因之血漿DNA之胎盤貢獻 大致10% 不顯著(接近零) 攜有N等位基因之血漿DNA之胎盤貢獻 大致10% 大致20% M及N等位基因之胎盤貢獻的比率(M:N) 1:1 0:2 表 4 . 針對NN父體基因型的M及N等位基因之胎盤貢獻。 在表4中，可比較M及N等位基因之百分比胎盤貢獻。兩種等位基因之大致相同胎盤貢獻(例如在彼此之臨限值內)表明胎兒基因型為MN。另一方面，N等位基因相比於M等位基因顯著較高的胎盤貢獻將指示NN之胎兒基因型。 在另一實施例中，不需要顧及父體基因型。在此情況下，胎兒之可能的基因型包括MM、MN及NN。    胎兒基因型    MN NN MM 攜有M等位基因之血漿DNA之胎盤貢獻 大致10% 不顯著(接近零) 大致20% 攜有N等位基因之血漿DNA之胎盤貢獻 大致10% 大致20% 不顯著(接近零)  M及N等位基因之胎盤貢獻的比率(M:N) 1:1 0:2 2:0 表 5 . 針對未知父體基因型的M及N等位基因之胎盤貢獻。 在表5中，顯示針對不同胎兒基因型的攜有M及N等位基因之DNA片段之胎盤貢獻。當胎兒具有MM之基因型時，M等位基因之胎盤貢獻將顯著高於N等位基因之胎盤貢獻。當胎兒具有NN之基因型時，N等位基因之胎盤貢獻將顯著高於N等位基因之胎盤貢獻。當胎兒具有NM之基因型時，M等位基因之胎盤貢獻將大致等於N等位基因之胎盤貢獻。 因此，當未知父體基因型時，可測定兩種等位基因之百分比貢獻。亦即，第一百分比貢獻可使用對準至基因座且包括N之第一組游離之DNA分子測定。第一組游離之DNA分子之甲基化程度可在接近基因座之K基因組位點量測。且第二百分比貢獻可使用對準至基因座且包括M之第二組游離之DNA分子測定。第二組游離之DNA分子之甲基化程度可在接近基因座之K基因組位點量測。對於胎兒基因型為MN之第一情形，針對任一等位基因測定之百分比貢獻將大致相同，如可藉由測定百分比貢獻是否在彼此之臨限值內檢驗。 為了說明此方法之可行性，吾人分析懷孕女性之血漿DNA。血漿DNA經亞硫酸氫鹽轉化且使用大規模平行定序進行分析。另外，分析胎盤及血細胞以測定胎兒及母親之基因型。出於說明的目的，分析位於KLF2 基因內之SNP。對於此SNP，母親及胎兒之基因型分別為CG及CC。在此基因型組合下，胎盤將C等位基因貢獻於母體血漿，但母體血漿中之所有G等位基因將衍生自母體組織。 在定序資料中，存在24個攜有G等位基因之片段及55個攜有C等位基因之片段。此等DNA片段內之CpG位點用於甲基化解卷積。在此分析中，一個目標為測定兩種等位基因之胎盤貢獻。為了說明原理，僅將胎盤及血細胞視為甲基化解卷積分析之候選組織。在另一實施例中，三種或大於三種類型之組織可用作候選物。在另一實施例中，組織預期具有顯著貢獻，例如血細胞、肝臟、肺、腸及胎盤可用作候選物。    C等位基因 G等位基因 C/G比 胎盤 62.6% 1.8% 34 血球 37.4% 98.2%    表 6 . 針對未知父體基因型的C及G等位基因之胎盤貢獻。 在表6中，來自胎盤之貢獻對於C等位基因及G等位基因分別推論為62.6%及1.8%。C/G之胎盤貢獻比為34。此等結果表明胎兒之基因型將為CC。此與胎盤組織之基因分型結果一致。 此實施例不同於基於具有特定甲基化模式之DNA之等位基因比率分析的非侵入性產前測試之前述方法且潛在地更具效用(Tong等人 Clin Chem 2006; 52: 2194-202)。在此前述方法中，組織特異性DNA首先基於甲基化模式鑑別自DNA混合物(例如血漿DNA)。舉例而言，特定基因在血細胞中完全未甲基化且在胎盤中經甲基化。鑑別使用使甲基化胎盤DNA保持完整之酶進行。 因此，血漿中剩餘的所有甲基化DNA分子將衍生自胎盤而非衍生自血細胞。接著，位於胎盤衍生之DNA分子上之SNP之等位基因比率可藉由使用完整胎盤DNA量測基因座處之不同等位基因之量測定。當胎兒對於SNP為雜合時，胎盤特異性DNA中之兩種等位基因的比率將為大致1。然而，若胎兒受非整倍體染色體影響且具有三個攜有此特定SNP之染色體複本，則兩種等位基因的比率將為1:2或2:1。 在此前述方法中，組織特異性DNA分子需要首先基於所關注的組織特有的甲基化狀態進行鑑別。甲基化DNA分子對於胎盤獨特，因為就目標區域而言，血細胞完全未甲基化。然而，在本發明實施例中，不需要某一甲基化狀態之獨特性。候選組織僅需要在其甲基化圖譜中不同，因此可使用更多基因座，進而實現單倍型解卷積。因此，可基於不同等位基因之甲基化圖譜測定其組織貢獻。另外，前述方法可更易受統計變化，因為各等位基因之情況下之胎兒讀數數目直接彼此相比。然而，當胎盤貢獻彼此相比時，胎兒讀數數目不直接彼此相比。取而代之，胎盤貢獻測定自所有讀數(甲基化或非甲基化)，且因此胎盤貢獻可相同，即使在胎兒讀數數目不同時亦如此。因此，可導致對一種單倍型之覆蓋度偏差。B . 使用解卷積測定遺傳單倍型 先前已展示經由懷有胎兒之懷孕女性之血漿DNA(或其他游離DNA)之分析，可使用相對單倍型劑量分析(RHDO)方法推論由胎兒遺傳之母體單倍型(Lo等人 Sci Transl Med 2010; 2: 61ra91及美國專利8,467,976)。在此方法中，吾人對於懷孕女性使用單倍型資訊。可使用家族分析或直接分析單倍型之方法(例如Fan等人 Nat Biotechnol 2011; 29: 51-57；Snyder等人 Nat Rev Genet 2015; 16: 344-358)獲得此後一資訊。在母親中雜合但在父親中純合之SNP可用於RHDO分析。此類使用特定SNP可限制可使用的基因座，且因此限制資料及精確性的量。實施例可不如此受限於此類特定SNP。另外，實施例可與以上參考組合使用以提供增加的精確性。 實施例可使用甲基化解卷積以使用游離之DNA分子對於兩種單倍型測定胎盤貢獻。胎盤貢獻可經比較以測定由胎兒遺傳何種單倍型。實施例可以推論之母體或父體單倍型起始，且接著量測彼等推論之單倍型中之每一者中含有SNP等位基因之血漿DNA分子之甲基化程度。吾人可隨後進行甲基化解卷積。胎兒單倍型可鑑別為具有來自甲基化解卷積分析之最高胎盤貢獻之單倍型。在所有以上實施例中，父體或母體單倍型亦可藉由家族分析(亦即藉由分析其他家族成員之DNA)或藉由直接方法(例如由Fan等人 Nat Biotechnol 2012描述之方法)測定，而非推論之單倍型。 1. 母體單倍型 在此實例中，吾人展示可用於推論由未出生的胎兒遺傳之母體單倍型之血漿DNA甲基化解卷積分析。來自懷孕女性之基因組DNA源，例如白血球層DNA可經受基因分型，例如使用微陣列。接著，將母體基因分型結果輸入單倍型推論程式(例如IMPUTE2，Howie等人 PLoS Genet. 2009;7:e 1000529)中以推論可能的第一母體單倍型及第二母體單倍型。群體特異性基因型及單倍型資訊可考慮在內以改良推論之精確性。在其他實施例中，親體單倍型可藉由單一分子分析，例如(但不限於)由Fan等人(Nat Biotechnol. 2011 ;29:51-7)、Kaper等人(Proc Natl Acad Sci USA. 2013; 110:5552-7)、Lan等人(Nat Commun. 2016;7:11784)及Selvaraj等人(Nat Biotech 2013;31:1111-1118)描述之方法產生。接著，母體血漿DNA可經受全基因組亞硫酸氫鹽定序及與參考基因組序列之比對。可隨後對於預測之單倍型中之每一者進行甲基化解卷積。由於母體血漿中之胎兒DNA主要源自胎盤，由胎兒遺傳之母體單倍型為顯示最高胎盤貢獻之單倍型。 母體單倍型資訊可用於將SNP等位基因及相同同源染色體上之CpG位點連接在一起。接著，可使用SNP等位基因鑑別來自相同染色體複本(單倍型)之DNA片段。此特定染色體複本(單倍型)上之CpG位點(或其他位點)可用於甲基化解卷積。由於可用於解卷積之CpG位點數目將與同源染色體上之SNP數目成比例且比基於單倍型之解卷積分析中連接至單一SNP之CpG位點數目大得多，此方法將比使用連接至單一SNP之CpG位點之解卷積分析更精確。原理說明於圖12A中。 圖12A顯示根據本發明之實施例使用甲基化解卷積由胎兒遺傳之母體單倍型之測定。在圖12A之頂部部分中，母親及胎兒之兩種單倍型顯示於三個基因座處且母親為雜合的。兩種母體單倍型標記為Hap I及Hap II。在此實例中，胎兒自母親遺傳Hap I。出於說明目的，僅顯示母親為雜合之SNP基因座。出於說明目的，在此實例中，父親對於此等基因座中之每一者為純合的。然而，相同原理擴展至父親在無任何變化的情況下為雜合之情形。 在圖12A之底部部分(經標記母體血漿)中，顯示各基因座處之兩個等位基因之實例，其中各實例表示所關注的基因座處之血漿中之不同DNA分子。出於說明的目的，僅顯示少數DNA分子。在此實例中，胎兒DNA百分比假設為20%，如藉由20%之DNA分子用P標記所示。 在母體血漿中，攜有胎兒基因型之DNA分子衍生自胎盤且因此攜有胎盤特異性甲基化模式。用「P」標記之圓形表示接近雜合基因座之展現胎盤甲基化模式之CpG位點。包括雜合基因座及相鄰位點之讀數可用於量測甲基化程度以偵測胎盤甲基化模式。在此實例中，一個目標為測定胎兒是否自母親遺傳Hap I或Hap II。為達成此目標，選擇在Hap I上攜有等位基因且覆蓋至少一個CpG位點之血漿DNA片段用於甲基化解卷積。由於胎兒自母親遺傳Hap I，胎盤將向此池貢獻顯著比例之血漿DNA分子。另一方面，當藉由甲基化解卷積分析在Hap II上攜有等位基因之片段時，將觀測到自胎盤之極低貢獻。 為了說明此情形，吾人分析關於表6在上文陳述之母體血漿樣品。吾人聚焦於染色體1上之5 Mb區域。選擇母親為雜合且胎兒為純合之SNP用於分析。對於此等SNP基因座中之每一者，在母親與胎兒之間共用之等位基因形成一個單倍型(表示為Hap I)且僅存在於母體基因組上之等位基因形成另一單倍型(表示為Hap II)。因此，在此實例中，存在兩種母體單倍型(Hap I及Hap II)且胎兒自母親遺傳Hap I。在母體血漿中，在Hap I上攜有等位基因之DNA片段及在Hap II上攜有等位基因之彼等分別使用甲基化解卷積分析。雜合SNP之相同血漿DNA分子上之所有CpG位點用於解卷積分析。在此實例中，此等CpG位點中無一者與I型或II型標記物重疊。    Hap I Hap II 肝臟 0% 0% 肺 0% 6.7% 結腸 3.4% 6.2% 小腸 0% 10.6% 胰臟 4.1% 0% 腎上腺 0% 4.6% 食道 0% 0% 脂肪組織 3.7% 3.6% 心臟 0% 0% 大腦 6.8% 10.6% T細胞 6.8% 21% B細胞 8.9% 11.7% 中性粒細胞 12.7% 25% 胎盤 53.5% 0% 表 7 . Hap I及Hap II之甲基化解卷積。 表7顯示在兩種母體單倍型，即Hap I及Hap II上攜有等位基因之血漿DNA片段之解卷積。胎兒遺傳母體Hap I。自此解卷積分析，胎盤推論為貢獻53.5%在Hap I上攜有等位基因之血漿DNA片段。另一方面，不存在胎盤對在Hap II上攜有等位基因之血漿DNA片段之貢獻。因此，甲基化解卷積分析已精確預測胎兒自母親遺傳Hap I。可使用與I型及/或II型標記物重疊之CpG位點達成較大精確性。 作為另一實例，為了展示此方法之實際效用，募集另一懷孕女性。獲取母體外周血。血液樣品分離為血漿及細胞組分。使用Illumina HumanOmni2.5-8 BeadChip陣列分析母體白血球層。吾人使用IMPUTE2 (Howie等人 PLoS Genet. 2009;7:e1000529)推論染色體1p之端粒末端上之5 Mb區域上之851個雜合SNP之相。單倍型定相係基於1000個基因組之參考單倍型(mathgen.stats.ox.ac.uk/impute/1000GP_Phase3.tgz)。 在獲得定相單倍型之後，連接至兩種單倍型之CpG位點用於進行甲基化解卷積。雜合SNP之相同血漿DNA分子上之所有CpG位點用於解卷積分析。在此實例中，此等CpG位點中無一者與I型或II型標記物重疊。在851個用於解卷積之SNP中，820個(96.2%)在內含子及基因間區域上。其中無一者與CpG島或岸重疊。    Hap I Hap II 肝臟 0 0 肺 0 5.4 結腸 0 6.2 小腸 0 0 胰臟 0 25 腎上腺 0 0 食道 0 0 脂肪組織 0 17.8 心臟 0 0 大腦 0 0 T細胞 11 7.9 B細胞 0 0 嗜中性白血球 20.2 28.4 胎盤 68.9 9.3 表 8 . Hap I及Hap II之甲基化解卷積。 表8顯示自一組參考單倍型推論之兩種母體單倍型上攜有等位基因之血漿DNA片段之解卷積。兩種單倍型命名為Hap I及Hap II。推論之Hap I具有顯著高於Hap II之量的胎盤貢獻，即68.9%相對於9.3%。因此，推論母體Hap I已由胎兒遺傳。依賴於單倍型推論之母體遺傳與來自母體及胎兒基因型之結果一致。 此方法之優勢為不限於胎兒之父親為純合且胎兒之母親為雜合之SNP。實際上，在以上實例中，吾人已在不知曉或推論父體基因型或單倍型的情況下進行分析。此為優於上述方法之優勢((Lo等人 Sci Transl Med 2010; 2: 61ra91；美國專利8,467,976；Fan等人 Nature 2012; 487: 320- 324；Kitzman等人 Sci Transl Med 2012; 4: 137ra76)。 在一些實施例中，第一單倍型之第一百分比貢獻可相比於基於胎兒DNA百分比濃度衍生之參考值以測定單倍型是否由胎兒遺傳。閾值可計算為例如(但不限於)胎兒DNA百分比濃度之1倍、1.2倍、1.4倍、1.6倍、1.8倍、2倍、2.2倍、2.4倍、2.6倍或2.8倍。以此方式，若第一百分比貢獻足夠大，則不需要測定第二單倍型之第二百分比貢獻。 在一些實施例中，遺傳性單倍型可具有胎兒分率雙倍之解卷積分率濃度且非遺傳性單倍型具有不顯著貢獻。非遺傳性單倍型之貢獻可不具有零貢獻，因為父體單倍型可由於一些父體等位基因可與母體等位基因相同而對此分析給出噪聲。若噪聲程度高，則可測定第二單倍型之百分比貢獻，且具有較高解卷積分率之單倍型可推論為由胎兒遺傳。 一些實施方案可使用參考值測試兩種單倍型以證實僅遺傳一種。若似乎遺傳兩者，則兩個百分比貢獻可彼此相比。另外，若似乎遺傳兩者，則可檢驗父體基因組，因為胎兒可能遺傳匹配非遺傳之母體單倍型之父體單倍型。 在其他實施例中，第二百分比貢獻可用於測定參考值，例如第二百分比貢獻加上臨限值。因此，參考值可為第二百分比貢獻與臨限值之總和。2. 父體單倍型 在另一實施例中，甲基化解卷積分析可應用於父體單倍型遺傳之分析。 圖12B顯示根據本發明之實施例之父體單倍型甲基化分析之說明。甲基化解卷積可對在父體Hap III及Hap IV上攜有等位基因之母體血漿DNA片段進行。由於胎兒已遺傳Hap III，相比於Hap IV，Hap III之胎盤貢獻將較高。因此，可推論胎兒之父體遺傳。 此實施例具有優於前述基於父體特異性等位基因之分析之方法的優點。舉例而言，對於SNP位置1，A等位基因存在於父親中，但不存在於母親中。因此，母體血漿中之父體特異性A等位基因之偵測指示由胎兒遺傳Hap III。然而，對於位置2處之SNP，C及T等位基因均不為胎兒特異性的。在此情況下，無法使用父體特異性等位基因分析。然而，甲基化解卷積分析不需要存在父體特異性等位基因。因此，在父親及母親兩者中雜合之SNP可用於兩種父體單倍型之甲基化解卷積分析。 因此，如用於母體單倍型之類似方法可用於測定遺傳何種父體單倍型。在圖12B中，Hap III之胎盤貢獻將高於Hap IV之胎盤貢獻。父體單倍型可以可測定母體單倍型之相同或類似方式測定。 3. 使用解卷積之方法 圖13為說明使用根據本發明之實施例之甲基化解卷積測定來自母體樣品之胎兒基因組之一部分之方法1300的流程圖。生物樣品包括來自複數種組織類型，包括母體組織類型及胎兒組織類型之游離之DNA分子的混合物。胎兒具有父親及為懷孕女性之母親。胎兒基因組部分可為整個染色體複本或染色體複本之僅一部分。胎兒基因組之測定部分可經組合以提供關於胎兒基因組之不同部分，至多整個胎兒基因組之資訊。 在步驟1310處，分析來自生物樣品之複數個游離之DNA分子。步驟1310可使用圖1之方法100之步驟140中所述之技術進行。舉例而言，可分析至少1,000個游離之DNA分子以測定游離之DNA分子所位於之位置，且甲基化程度可如下所述地量測。另外，分析游離之DNA分子以測定游離之DNA分子之各別等位基因。舉例而言，DNA分子之等位基因可測定自獲自定序或獲自雜交至DNA分子之特定探針之序列讀數，其中兩種技術均可提供序列讀數(例如探針可在存在雜交時視為序列讀數)。 在步驟1320處，測定胎兒之第一親體之第一親體基因組之第一染色體區域之第一單倍型及第二單倍型。熟習此項技術者將瞭解測定親體之單倍型的各種技術。單倍型可測定自與用於測定下文甲基化程度相同的樣品或測定自不同樣品。在一些實施方案中，單倍型可測定自細胞樣品，例如血液樣品之白血球層或另一器官之組織。測定單倍型之實例提供於美國專利第8,467,976號中，其以全文引用的方式併入本文中。第一親體可為母親或父親。偵測親體單倍型之方法的其他實例包括(但不限於)由Fan等人(Nat Biotechnol 2011; 29: 51-57)、Snyder等人(Nat Rev Genet 2015; 16: 344-358)描述之方法、來自10X Genomics之GemCode技術(www.10xgenomics.com/)及來自Cergentis之靶向基因座擴增(TLA)技術(www.cergentis.com/)。 在步驟1330處，一或多個雜合基因座鑑別自第一及第二單倍型。各雜合基因座具有第一單倍型中之對應第一等位基因及第二單倍型中之對應第二等位基因。一或多個雜合基因座可為第一複數個雜合基因座，其中第二複數個雜合基因座可對應於不同染色體區域。 在步驟1340處，鑑別第一組複數個游離之DNA分子。複數個游離之DNA分子中之每一者位於來自步驟1330之雜合基因座中的任一者處且包括對應第一等位基因，以使得游離之DNA分子可鑑別為對應於第一單倍型。可能的是游離之DNA分子位於大於一個雜合基因座處，但通常，一個讀數將僅包括一個雜合基因座。第一組游離之DNA分子中之每一者亦包括N個基因組位點中之至少一者，其中基因組位點用於量測甲基化程度。N為整數，例如大於或等於2、3、4、5、10、20、50、100、200、500、1,000、2,000、或5,000。因此，游離之DNA分子之讀數可指示1位點、2位點等的覆蓋度。 在步驟1350處，使用第一組複數個游離之DNA分子量測N個基因組位點(例如CpG位點)處之N個第一混合物甲基化程度。可對於N個基因組位點中之每一者量測一個第一混合物甲基化程度。步驟1350可以與圖1之方法100之步驟150類似之方式進行。在一些實施例中，DNA分子之甲基化程度之量測可使用甲基化感測定序結果，其亦可用於測定DNA分子之位置及各別等位基因。熟習此項技術者將瞭解可用於測定DNA分子上之位點之甲基化狀態的各種技術。 在步驟1360處，使用N個第一甲基化程度測定混合物中之胎兒組織類型之第一百分比貢獻。在一些實施例中，步驟1360可經由圖1之方法100之步驟160及170進行。因此，可對於一組M種組織類型同時測定百分比貢獻。步驟1360可使用對於M種組織類型中之每一者測定之N個基因組位點之N個組織特異性甲基化程度，例如圖1之方法100之步驟120。 在步驟1370處，鑑別第二組複數個游離之DNA分子。複數個游離之DNA分子中之每一者位於來自步驟1330之雜合基因座中的任一者處且包括對應第二等位基因，以使得游離之DNA分子可鑑別為對應於第二單倍型。第二組游離之DNA分子中之每一者亦包括N個基因組位點中之至少一者，其中基因組位點用於量測甲基化程度。 在步驟1380處，使用第二組複數個游離之DNA分子量測N個基因組位點之N個第二混合物甲基化程度。步驟1380可以類似於步驟1350的方式進行。 在步驟1385處，使用N個第二甲基化程度測定混合物中之胎兒組織類型之第二百分比貢獻。步驟1385可以類似於步驟1360的方式進行。 在步驟1390處，計算第一百分比貢獻與第二百分比貢獻之間的第一分離值。分離值之實例描述於本文中，例如包括差值或比率。 在步驟1395處，基於第一分離值在一或多個雜合基因座處測定胎兒基因組之部分。因此，可測定第一親體之遺傳單倍型。舉例而言，第一分離值可為第一百分比貢獻與第二百分比貢獻的比率。當比率大於臨限值時，胎兒基因組之部分可測定為具有第一單倍型之一或多個複本且無第二單倍型之複本。臨限值之實例包括(但不限於)1.3、1.4、1.5、1.6、1.8、2.0、2.2、2.4、2.6、2.8及3.0。當比率小於臨限值時，胎兒基因組之部分可測定為具有第二單倍型之一或多個複本且無第一單倍型之複本。臨限值之實例包括(但不限於)0.1、0.2、0.3、0.4、0.5、0.6、0.7及0.8。當比率等於閾值內之一者時，胎兒基因組之部分可測定為具有第一單倍型及第二單倍型。閾值之實例包括(但不限於)0.85、0.9、0.95、1.0、1.05、1.1及1.15。當兩個親體在分析之區域中具有相同單倍型時，可遺傳兩種單倍型。 作為另一實例，第一分離值為第一百分比貢獻與第二百分比貢獻之差值。當差值大於臨限值時，胎兒基因組之部分可測定為具有第一單倍型之一或多個複本且無第二單倍型之複本。臨限值之實例包括(但不限於)1%、1.5%、2%、2.5%、3%、4%、5%、6%、7%、8%、10%、12%、14%、16%、18%及20%。當差值小於臨限值，例如當臨限值為負數時，胎兒基因組之部分可測定為具有第二單倍型之一或多個複本且無第一單倍型之複本。 亦可測定另一親體之遺傳單倍型。舉例而言，可在另一親體之基因組中鑑別第一染色體區域之第二複數個雜合基因座。可對於另一親體之單倍型中之每一者測定百分比貢獻，且分離值可用於測定另一親體之遺傳單倍型。 舉例而言，第一複數個雜合基因座及第二複數個雜合基因座可為相同基因座或不同。第二複數個雜合基因座中之每一者可包括另一親體之第一單倍型(例如第一父體單倍型)組之對應第三等位基因及另一親體之第二單倍型(例如第二父體單倍型)中之對應第四等位基因。第三及第四等位基因可與第一及第二等位基因相同。除第一親體之第一及第二組游離之DNA分子以外，第三組複數個游離之DNA分子可各位於第二複數個雜合基因座中的任一者處，包括雜合基因座之對應第三等位基因，且包括K個基因組位點中之至少一者。K個基因組位點可與用於第一親體之N個基因組位點相同或不同。以與第一親體類似之方式，可使用第三組的第二複數個游離之DNA分子在K個基因組位點量測K第三混合物甲基化程度，且可使用K第三甲基化程度測定混合物中之胎兒組織類型之第三百分比貢獻。第三百分比貢獻對應於另一親體之第一單倍型(例如第一父體單倍型)。 第四組複數個游離之DNA分子可各位於第二複數個雜合基因座中的任一者處，包括雜合基因座之對應第四等位基因，且包括K個基因組位點中之至少一者。因此，第四組DNA可用於測試另一親體之第二單倍型。可使用第四組的第二複數個游離之DNA分子量測K個基因組位點之K第四混合物甲基化程度，且可使用K第四甲基化程度測定混合物中之胎兒組織類型之第四百分比貢獻。可計算第三百分比貢獻與第四百分比貢獻之間的第二分離值，且可基於第二分離值測定第二複數個雜合基因座處之胎兒基因組之部分。來自另一親體之遺傳單倍型可以與用於第一親體類似之方式測定。第四百分比貢獻對應於另一親體之第二單倍型(例如第二父體單倍型)。 在一些實施例中，不需要測定第二百分比貢獻。取而代之，若對應百分比貢獻足夠高，則單倍型可測定為遺傳的。舉例而言，第一百分比貢獻可相比於參考值以測定胎兒是否在第一染色體區域遺傳第一單倍型。當第一百分比貢獻超過參考值時，胎兒可測定為在第一染色體區域遺傳第一單倍型。 在其他實施例中，參考值可測定自第二百分比貢獻。舉例而言，參考值可為第二百分比貢獻與臨限值之總和。與臨限值之總和可確保第一百分比貢獻充分地大於第二百分比貢獻。 可藉由比較第二百分比貢獻與參考值對第二單倍型進行遺傳之分開測定以測定胎兒是否在第一染色體區域遺傳第二單倍型。當第二百分比貢獻超過參考值時，胎兒可測定為在第一染色體區域遺傳第二單倍型。若兩個百分比貢獻均測定為超過參考值，則兩個百分比貢獻可彼此相比以測定一者是否顯著大於另一者(例如使用臨限值)。可測定另一親體之單倍型以鑑別此等單倍型中的一者是否與第一親體之單倍型相同，進而解釋可遺傳第一親體之兩種單倍型。C . 使用甲基化程度測定遺傳單倍型 其他實施例可使用游離胎兒DNA之一般低甲基化將遺傳單倍型鑑別為具有較低總甲基化程度之一者。實施例可以推論之母體或父體單倍型起始，且接著量測彼等推論之單倍型中之每一者中含有SNP等位基因之血漿DNA分子之甲基化程度。在分析母體單倍型之一個實施方案中，可比較兩個推論母體單倍型之甲基化程度，且具有較低甲基化程度之一者將預測為由胎兒遺傳之單倍型。在分析父體單倍型之另一實施方案中，可比較兩個推論父體單倍型之甲基化程度，且具有較低甲基化程度之一者將預測為由胎兒遺傳之單倍型。1. 實例 舉例而言，可測定兩種母體單倍型中之每一者之甲基化程度。由於胎盤組織相比於其他組織相對低甲基化，吾人預期由胎兒遺傳之母體單倍型將比不由胎兒遺傳之單倍型更低甲基化。使用母親之實際單倍型在母體血漿中測試甲基化密度，該等單倍型使用母體、父體及胎兒基因型推論。    Hap I Hap II 總甲基化密度 65% 87% 表 9 . 實際Hap I及Hap II之甲基化密度。 表9顯示母體血漿中之兩種母體單倍型之甲基化密度。由於Hap I為根據基因分型由胎兒遺傳之實際單倍型，單倍型之甲基化分析結果恰當地鑑別遺傳。 在其他實施例中，可基於單獨的母親基因型推論母體單倍型，或來自單倍型資料庫之群體之參考單倍型亦可用於此分析。用於此實例之母體單倍型使用IMPUTE2程式定相。因此，推論之母體單倍型亦可用於此分析。    Hap I Hap II 總甲基化密度 68% 76% 表 10 . 推論之Hap I及Hap II之甲基化密度。 表10顯示母體血漿中之兩種推論之母體單倍型之甲基化密度。由胎兒遺傳之推論之母體單倍型具有較低甲基化密度。可用於測定一種單倍型是否具有足夠較低的甲基化密度之統計程序之實例包括卡方檢驗(chi-square test)。可能需要兩個甲基化程度之間的分離足夠大(例如大於臨限值)以進行測定。若分離不足，則可進行不確定分類。在一些實施例中，兩種單倍型之遺傳之測定可測定分離是否不足夠大且兩個甲基化程度是否均低於臨限值水準，其可藉由包括胎兒DNA表徵。舉例而言，表9及10指示低於70%之甲基化密度可指示胎兒遺傳該單倍型。當親體對於分析之區域共用單倍型時，可遺傳兩種單倍型。 在另一實施例中，可比較攜有父體Hap III及Hap IV之母體血漿DNA之總甲基化密度。與母體單倍型分析類似，胎兒將推論為遺傳具有較低總甲基化密度之父體單倍型。2. 使用甲基化程度之方法 圖14為說明使用根據本發明之實施例之甲基化程度測定來自母體樣品之胎兒基因組之一部分之方法1400的流程圖。生物樣品包括來自複數種組織類型，包括母體組織類型及胎兒組織類型之游離之DNA分子的混合物。胎兒具有父親及為懷孕女性之母親。胎兒基因組部分可為整個染色體複本或染色體複本之僅一部分。胎兒基因組之測定部分可經組合以提供整個胎兒基因組，如同本文所述之其他方法。 在步驟1410處，分析來自生物樣品之複數個游離之DNA分子。步驟1410可以與圖13之方法1300之步驟1310類似之方式進行。 在步驟1420處，測定胎兒之第一親體之第一親體基因組之第一染色體區域之第一單倍型及第二單倍型。步驟1420可以與圖13之步驟1320類似之方式進行。在一些實施例中，可使用來自第一親體之樣品，例如血液樣品或可包含或可不包含胎兒DNA之其他組織在複數個雜合基因座處測定第一親體之基因組之基因型。可獲得複數個參考單倍型，例如自參考基因組之資料庫。可使用基因型及複數個參考單倍型推論第一單倍型及第二單倍型。舉例而言，可相對於參考單倍型比較各基因型之等位基因，且可丟棄在對應基因座處不包括等位基因之任何單倍型。一旦保留兩個參考單倍型，彼等單倍型可鑑別為第一單倍型及第二單倍型。 在步驟1430處，自第一及第二單倍型鑑別複數個雜合基因座。各雜合基因座具有第一單倍型中之第一等位基因及第二單倍型中之第二等位基因。 在步驟1440處，鑑別第一組複數個游離之DNA分子。步驟1440可以與圖13之步驟1340類似之方式進行。 在步驟1450處，使用第一組複數個游離之DNA分子量測第一混合物甲基化程度。舉例而言，第一混合物甲基化程度可為第一組游離之DNA分子之甲基化密度。甲基化密度可計算為第一組之所有游離之DNA分子之總甲基化密度。在另一實例中，可對於各基因座計算分離甲基化密度，且分離甲基化密度可經組合以獲得第一混合物甲基化程度，例如分離甲基化密度之平均值。 在步驟1460處，鑑別第二組複數個游離之DNA分子。1460可以與圖13之步驟1370類似之方式進行。 在步驟1470處，使用第二組複數個游離之DNA分子量測第二混合物甲基化程度。舉例而言，第二混合物甲基化程度可為第二組游離之DNA分子之甲基化密度。 在步驟1480處，基於第一混合物甲基化程度及第二混合物甲基化程度中之何者較低測定由胎兒遺傳第一單倍型及第二單倍型中之何者。作為步驟1480之部分，可在第一混合物甲基化程度與第二混合物甲基化程度之間測定分離值，且相比於臨限值。臨限值可確保較低水準足夠低。臨限值可使用卡方檢驗測定。舉例而言，可對已知遺傳單倍型之樣品進行量測，且可測定分離值之分佈，且可選擇精確測定獲自樣品之訓練資料中之遺傳單倍型之臨限值。方法1300及1400亦可組合，其中各方法以檢查形式進行，且遺傳單倍型測定兩種方法是否彼此一致。D . 基因座之選擇 各種實施例可用於比較母體血漿中兩種推論母體單倍型之甲基化程度或百分比貢獻。在一個實施例中，可在分析之前測定欲分析之SNP基因座數目。舉例而言，可根據多種因素，例如(但不限於)所需統計檢定力、所關注區域中胎盤及血球之甲基化程度平均差異及對於各SNP分析之分子數目，測定用於單倍型解卷積分析之SNP基因座數目。 所關注區域之尺寸可為固定的，且所關注區域內之所有SNP可用於分析。可考慮多種因素，例如(但不限於)所需統計檢定力、所關注區域中胎盤及血球之甲基化程度平均差異、對於各SNP分析之分子數目及與所關注區域減數分裂重組之機率，來測定所關注區域之尺寸。 在其他實施例中，在分析之前不測定SNP數目及欲分析區域的尺寸。舉例而言，SNP數目可依序增加直至資料足以獲得關於何種母體單倍型比另一者在統計學上顯著地較少甲基化之統計學上顯著結論。舉例而言，所關注區域上之SNPs可以其基因組座標之升序排列。接著，可用具有基因組座標之最低數目的SNP之資料進行統計測試。若此足以作出關於何種單倍型在統計學上較少甲基化之結論，則作出結論。類似地，SNPs可以足夠使用的基因組座標之最高數目降序排列。 若統計精確性不足，則可以具有基因組座標之較高數目的下一SNP為起始物進行另一統計比較。另一方面，若第一SNP之資料不足以得出一種單倍型比另一種較少甲基化(或百分比貢獻之間的分離值不足夠大)之結論，則可添加另一SNP之資料且進行另一輪統計測試。可繼續此程序直至累積資料足以作出統計顯著結論。可進行多個統計測試以比較兩種單倍型之甲基化程度，例如(但不限於)斯圖登氏t檢驗(Student's t-test)、曼-惠氏秩和檢驗(Mann-Whitney rank-sum test)及卡方檢驗。可基於結論之所需置信度測定統計顯著性程度，例如(但不限於)採用0.05、0.01、0.001、0.0001或0.00001之P-值。E . 與 RHDO 之 組合 在一些實施例中，由美國專利8,467,976之RHDO分析產生的結果可與本發明甲基化實施例組合以達成較精確診斷程序或減少所需的定序量。舉例而言，可使用本發明實施例及使用美國專利8,467,976之RHDO分析之結果測定胎兒單倍型，且可比較測定自兩種技術之胎兒單倍型。舉例而言，兩個分析之結果將僅在其一致時為可接受的。若兩個分析顯示不同結論，則可進行其他分析，例如可以較高深度之覆蓋度對基因組重複量測。 為使此類組合方法最具成本效益，較佳具有可對於兩種方法產生資料之一種類型的定序。在一個實施例中，此可藉由將產生定序以及甲基化資訊之單分子方法進行，例如使用來自Pacific Biosciences之單分子實時定序技術，或奈米孔定序(例如來自Oxford Nanopore Technologies)。此等為甲基化感測定序之兩個實例。在另一實施例中，可對亞硫酸氫鹽定序結果進行RHDO分析。對於此類實施例，亦可使用亞硫酸氫鹽定序測定任何母體及父體基因資訊。亞硫酸氫鹽定序因此為甲基化感測定序之另一實例。此外，可使用其他甲基化感測定序技術，如氧化亞硫酸氫鹽定序(Booth等人 Science 2012; 336: 934-937)或Tet輔助亞硫酸氫鹽定序(Yu等人 Cell 2012; 149: 1368-1380)。後者實例將允許吾人分析所分析DNA分子之5-甲基胞嘧啶分佈。F . 使用胎兒基因組之知識 胎兒基因組之非侵入性產前分析可用於測定胎兒是否自親體遺傳疾病。此特別適用於偵測單基因性疾病，例如先天性腎上腺增生(New等人 J Clin Endocrinol Metab 2014;99:E1022-30)、β-地中海型貧血(Lam等人 Clin Chem. 2012;58:1467-75)及遺傳性肌營養不良(Genet Med 2015;17:889-96)。若偵測到單基因性疾病，則可進行各種處理，例如可終止懷孕、在懷孕之前或出生之後提供之處理。舉例而言，類固醇處理可在產前向確認具有受先天性腎上腺增生影響之胎兒的懷孕女性給予以避免異常性發育。VI . 非整倍性偵測之單倍型解卷積分析 單倍型解卷積亦可用於偵測胎兒之染色體區域之序列不平衡，如非整倍體、微缺失或微擴增(例如微重複)。舉例而言，一個區域中之單倍型之百分比貢獻可相比於另一區域中之另一單倍型之百分比貢獻。A . 母親 圖15顯示對於根據本發明之實施例之母體單倍型基於單倍型解卷積之染色體非整倍性偵測。在此圖示中，母親具有兩個母體單倍型，即Hap I及Hap II。出於說明目的，吾人假定其血漿DNA之80%衍生自其自身細胞且20%衍生自胎盤，此為通常量測之範圍內的例示百分比。此方法可一般應用於具有不同胎兒DNA百分比之孕婦。不需要胎兒DNA百分比之知識，但僅提供用於說明，儘管胎兒DNA百分比之量測可以各種方法進行，例如使用胎兒特異性等位基因或胎兒特異性甲基化標記物。 胎兒遺傳Hap I且另一單倍型來自父親，即Hap III。胎盤衍生之DNA將展現胎兒基因型，且可藉由分析產生於胎盤衍生之DNA之百分比貢獻偵測序列不平衡。 如上文所說明，可經由兩種母體單倍型之解卷積測定母體單倍型之胎兒遺傳。可對於兩種母體單倍型中之每一者進行胎盤對母體DNA之貢獻的分析。相比於不由胎兒遺傳之母體單倍型(Hap II)，由胎兒遺傳之母體單倍型(在此實例中為Hap I)將具有高得多的胎盤貢獻。Hap I之胎盤貢獻將與母體血漿中之胎兒DNA百分比濃度正相關。 在測定由胎兒遺傳何種母體單倍型之後，可經由母體單倍型解卷積進一步測定胎兒自母親遺傳之染色體之劑量。在此圖示中，使用母體單倍型解卷積分析兩個染色體區域。在一個實施例中，參考染色體(RefChr)為不太可能受染色體非整倍性影響之染色體或染色體區域。參考染色體區域顯示於圖15之左側。目標染色體(TargetChr)為潛在地受染色體非整倍性影響之染色體或染色體區域。目標染色體區域顯示於圖15之右側。兩個區域可為相同染色體之不同區域或兩種不同染色體之區域。 在所示之實例中，已經由各區域之Hap I及Hap II之甲基化解卷積推論胎兒對於參考染色體及目標染色體兩者自母親遺傳Hap I。接著，可在參考染色體與目標染色體之間比較關於Hap I之母體血漿DNA之胎盤貢獻。若目標染色體區域之Hap I之胎盤貢獻顯著不同於參考染色體區域之Hap I之胎盤貢獻(例如由於擴增較高或由於缺失較低)，則可鑑別序列不平衡。 出於說明目的，吾人使用三染色體性之偵測作為實例。然而，亦可使用此方法偵測其他類型之染色體非整倍體(包括單染色體)、次染色體區域之擴增或次染色體區域之缺失。對於三染色體性，可自父親(表示為三染色體性(F))或母親(表示為三染色體性(M))遺傳受影響染色體之額外複本。在超過90%的第21對染色體三體症病例下，染色體21之額外複本衍生自母親(Driscoll等人 N Engl J Med 2009;360: 2556- 2562)。在三染色體(M)之情形下，目標染色體之Hap I之胎盤貢獻將高於參考染色體。在圖15中，三染色體(M)藉由Hap I之兩個實例顯示，其將對於目標區域提供比Hap I之一個實例對於參考區域高的胎盤貢獻。 可藉由比較兩個胎盤貢獻之間的分離值及臨限值(其可基於胎兒DNA百分比之分開量測)測定Hap I對於目標染色體之胎盤貢獻是否高於對於參考染色體之貢獻。較高胎兒DNA百分比將導致兩個胎盤貢獻之間的較高預期分離值且因此臨限值可設定為較高。舉例而言，在胎兒DNA百分比為20%之情況下，Hap I對於參考區域之胎盤貢獻將為約20%且Hap I對於目標區域之胎盤貢獻將為約36.4%。 舉例而言，假定10個DNA分子存在於參考染色體處，則其中之兩個為胎兒且其中之八個為母體。對於兩個胎兒DNA分子，一個衍生自Hap I且一個衍生自Hap III。對於八個母體DNA分子，四個為Hap I且四個為Hap II。對於目標區域，將存在來自胎兒之Hap I之額外DNA分子。因此，將總共存在兩個胎兒Hap I DNA分子及4個母體Hap I DNA分子，提供2/6=33.3%。差值(例如13.3%)之臨限值可置於0與13.3%之間以提供最優特異性及敏感性。分離值之分佈可測定自參考樣品組。在整倍體之情境下，胎盤貢獻將大致相同，例如分離值將小於臨限值。基於本文及美國專利8,467,976及本文中所引用之其他參考文獻中之描述，熟習此項技術者將知曉如何選擇適合臨限值。 在一個實施例中，各已知懷有整倍體胎兒之一群懷孕女性之目標與參考染色體之間的Hap I之胎盤貢獻之比率(或其他分離值)可用作參考間隔。測試案例中之比率可相比於此參考組以測定是否關於目標區域相對於參考區域存在Hap I之胎盤貢獻的顯著升高。在20%胎兒DNA的實例中，比率將為33.3/20=1.67。比率可一般為2/(1+f)，其中f表示胎兒DNA百分比濃度。在另一實施例中，可測定目標與參考染色體之間的Hap I之胎盤貢獻之差值。此差值接著相比於參考組。B . 父親 在另一實施例中，可在母體血漿中進行父體單倍型(Hap III及Hap IV)之單倍型解卷積。可以如關於母體單倍型類似之方式進行父體單倍型之分析。 圖16顯示對於根據本發明之實施例之父體單倍型基於單倍型解卷積之染色體非整倍性偵測。在此圖示中，父親具有兩個父體單倍型，即Hap III及Hap IV。如同圖15，胎兒自母親遺傳Hap I且自父親遺傳Hap III。 在染色體之額外複本衍生自父親(三染色體(F))之情形下，Hap III對於目標染色體之胎盤貢獻將高於對於參考染色體。此對於三染色體(F)實例顯示，其中顯示Hap III之兩個複本。如關於母體單倍型在上文所述，Hap III對於目標及參考區域之胎盤貢獻之間的分離值可相比於臨限值以測定目標區域是否存在Hap III之額外複本。在各種實施例中，測試案例之兩個胎盤貢獻之比率或差值可相比於各已知懷有整倍體胎兒之懷孕女性之參考組以測定胎兒是否具有目標染色體之染色體三染色體，或目標染色體區域之擴增或缺失。臨限值可基於整倍體胎兒之參考組、非整倍性胎兒之參考組或兩者之分離值。亦可使用胎兒DNA百分比之分開量測，如本文所述。C . 偵測序列不平衡之方法 圖17為根據本發明之實施例使用來自懷孕女性之生物樣品偵測懷孕女性之未出生胎兒之胎兒基因組之一部分中之序列不平衡之方法1700的流程圖。 在步驟1710處，分析來自生物樣品之複數個游離之DNA分子。步驟1710可以與圖13之方法1300之步驟1310類似之方式進行。 在步驟1720處，測定胎兒之第一親體之第一親體基因組之目標染色體區域的第一目標單倍型，且測定第一親體基因組之參考染色體區域之第一參考單倍型。步驟1720可以與圖13之步驟1320類似之方式進行。目標染色體區域及參考染色體區域可為整個染色體或染色體之僅一部分。因此，目標染色體區域可為第一染色體且參考染色體區域可為不同於第一染色體之第二染色體。第一親體可為胎兒之母親或父親。 目標染色體區域可基於各種標準選擇。舉例而言，可選擇複數個目標區域，因為可能想到測試多個指定尺寸之非重疊區域，如1 Mb、5 Mb、10 Mb、20 Mb、50 Mb等。作為另一實例，可基於將區域鑑別為具有比預期更多的DNA分子之複本數分析選擇目標染色體區域，例如如美國專利公開案2009/0029377及2011/0276277中所述。 在一些實施例中，可測定胎兒自第一親體遺傳第一目標單倍型及胎兒自第一親體遺傳第一參考單倍型。測定可包括圖13或圖14之實施例。舉例而言，測定胎兒自第一親體遺傳第一目標單倍型可包括測定對應於第二目標單倍型之混合物中之胎兒組織類型之第二目標百分比貢獻、計算第一目標百分比貢獻與第二目標百分比貢獻之間的第二分離值及基於第二分離值測定胎兒自第一親體遺傳第一目標單倍型。 在步驟1730處，鑑別第一親體基因組之目標染色體區域之複數個目標雜合基因座。各目標雜合基因座包括第一親體基因組之第一染色體區域之第一目標單倍型種之對應第一目標等位基因及第二目標單倍型中之對應第二目標等位基因。返回參看圖15之實例，目標雜合基因座具有Hap I上之{G,T,A}之對應第一目標等位基因且具有Hap II上之{A,G,C}之對應第二目標等位基因。 在步驟1740處，鑑別目標組之複數個游離之DNA分子。目標組之各游離之DNA分子位於目標雜合基因座中的任一者處，包括對應第一目標等位基因，且包括目標染色體區域中之N個基因組位點中之至少一者。步驟1740可以與本文所述類似之方式進行。舉例而言，序列讀數可映射至參考基因組，其中目標組之複數個游離之DNA分子與目標雜合基因座中的任一者對準。 在步驟1750處，使用目標組之複數個游離之DNA分子量測N個基因組位點之N個第一混合物甲基化程度。步驟1750可以與圖13之步驟1350類似之方式進行。 在步驟1760處，使用N個第一甲基化程度測定混合物中之胎兒組織類型之第一百分比貢獻。步驟1760可以與圖13之步驟1360類似之方式進行。 在步驟1770處，對於第一親體基因組之參考染色體區域鑑別複數個參考雜合基因座。各參考雜合基因座包括第一親體基因組之參考染色體區域中之第一參考單倍型中之對應第一參考等位基因及第二參考單倍型中之對應第二參考等位基因。返回參看圖15之實例，參考雜合基因座具有Hap I上之{A,T,C}之對應第一目標等位基因且具有Hap II上之{T,C,A}之對應第二目標等位基因。 在步驟1775處，鑑別參考組之複數個游離之DNA分子。參考組之各游離之DNA分子位於參考雜合基因座中的任一者處，包括對應第一參考等位基因，且包括參考染色體區域中之K個基因組位點中之至少一者。 在步驟1780處，K參考混合物甲基化程度使用複數個游離之DNA分子之參考組量測於K個基因組位點。 在步驟1785處，使用K參考甲基化程度在混合物中測定胎兒組織類型之第一參考百分比貢獻。 在步驟1790處，計算第一目標百分比貢獻與第一參考百分比貢獻之間的第一分離值。 在步驟1795處，第一分離值相比於臨限值以確定胎兒是否對於目標染色體區域具有序列不平衡之分類。若第一分離值超過臨限值，則可鑑別序列不平衡。可如上文所述測定臨限值，例如基於不具有序列不平衡之樣品之參考組及/或具有序列不平衡之樣品之參考組中可見之分離值。作為實例，分類可對於測試之序列不平衡為正性、負性或不確定的。 取決於序列不平衡的類型，可使用不同臨限值。舉例而言，若序列不平衡為缺失，則第一分離值將預期為負值。在此情況下，臨限值可為負數，且該比較可根據為較大負數測定第一臨限值超過臨限值。若序列不平衡測試為擴增，則其可測試分離值是否大於臨限值。因此，使用之臨限值可取決於測試之序列不平衡的類型。VII . 鑑別患病組織之簽名之解卷積 若已知基因組簽名(例如特定SNP等位基因)，則實施例可測定何種組織為此類簽名之來源。由於展現簽名之游離之DNA分子來自源組織，源組織可鑑別自使用展現簽名之游離之DNA分子測定之百分比貢獻。因此，具有移植器官之簽名(例如移植器官之單倍型之簽名)之游離之DNA分子可用於在高敏感度下監測來自移植器官之游離之DNA分子之量的變化，例如鑒於混合物中之DNA之高百分比貢獻將來自移植器官。對於移植提供實例以顯示技術為精確的。在另一實例中，腫瘤之簽名可用於鑑別腫瘤所駐留之組織。A . 器官移植 作為器官移植之實例，吾人分析接受肝臟移植之患者及接受骨髓移植之患者的血漿。對於各情況，經由來自患者及供體之組織的基因分型鑑別供體特異性SNP等位基因。對於肝臟移植受體，供體肝臟之生檢及受體之血細胞經定序。對於骨髓移植案例，頰黏膜拭子(受體基因型)及血細胞(供體基因型)經定序。血漿DNA樣品在亞硫酸氫鹽轉化之後定序。攜有供體特異性SNP等位基因及至少一個CpG位點之經定序DNA片段用於下游甲基化解卷積分析。對於接受肝臟及骨髓移植之患者分別定序總共7千2百萬及1.21億個讀數。對於兩種情況，38及5355個片段分別用於解卷積分析。 組織類型 肝臟移植受體 骨髓移植受體 肝臟 45.4 4.4 肺 0.0 1.5 結腸 29.3 6.3 小腸 0.0 1.8 胰臟 0.0 0.0 腎上腺 0.0 0.0 食道 0.0 0.0 脂肪組織 0.0 14.8 心臟 0.0 0.0 大腦 14.5 9.6 T細胞 0.0 12.3 B細胞 5.9 16.6 嗜中性白血球 4.9 32.8 表 6 . 兩個移植受體中不同器官對攜有供體特異性等位基因之血漿DNA片段之百分比貢獻。 表6顯示對肝臟移植受體及骨髓移植受體中攜有供體特異性等位基因之血漿DNA片段的甲基化解卷積分析。數目表示不同組織對供體特異性血漿DNA片段之百分比貢獻。對於肝臟移植案例，肝臟展示為此等DNA片段之最重要貢獻者。對於骨髓移植案例，造血系統(包括T細胞、B細胞及嗜中性白血球)為供體特異性DNA片段之主要貢獻。此等結果指示甲基化解卷積可精確指示具有單一核苷酸改變之DNA片段之組織來源。少量定序片段很可能由於量測不精確性歸因於其他組織，因為相對較小數目的供體特異性片段用於解卷積分析。 可以上述方式測定及監測與移植器官相關之組織的百分比貢獻。在基線百分比貢獻(參考百分比貢獻之實例)由於僅使用展現供體簽名之游離之DNA分子而相對較高之情況下，可偵測到血漿中之供體DNA之總量的較小變化。因此，甲基化解卷積分析可應用於監測器官移植。 如關於肝臟移植在上文可見，甲基化解卷積並非絕對特異性的。在此分析中，攜有供體特異性等位基因之血漿DNA片段用於甲基化解卷積分析。此等片段特異於供體且應僅衍生自此肝臟移植受體之肝臟。因此，肝臟之理論貢獻應為100%。另一可能性為某些細胞類型存在於不同類型的組織中，使得肝臟甲基化圖譜與其他組織重疊。舉例而言，肝臟中之結締組織細胞亦可存在於其他器官中。但可鑑別來自其他患者或本發明患者之其他樣品(例如在其他時間)之相對百分比是否釋放更多游離之DNA分子。 在各種實施例中，供體簽名可對應於供體基因組之特定單倍型或染色體區域中之兩種單倍型。甲基化解卷積可使用位於特定供體單倍型上之游離之DNA分子進行，且可監測特定單倍型之百分比貢獻之增加。若出現顯著增加(例如如藉由百分比或絕對臨限值所量測)，則可鑑別移植器官之排斥反應。 圖18顯示根據本發明之實施例用於器官移植監測之單倍型解卷積之圖示。供體具有標記為Hap I及Hap II之單倍型，且受體具有標記為Hap III及Hap IV之單倍型。供體具有基因座1及基因座3處之簽名，因為等位基因未發現於受體單倍型上。基因座2及基因座4不具有供體簽名。因此，實施例可使用位於基因座1及基因座3處之DNA分子作為解卷積方法之部分。 血漿DNA解卷積可用於測定來自移植器官之經測定百分比貢獻在基線處或相對於基線增加。在一些實施例中，若存在不同簽名，則可對於Hap I及Hap II中之每一者分別測定百分比貢獻；此類不同簽名可存在於不同基因座處。在其他實施例中，可對於兩種單倍型測定單一百分比貢獻，例如當其共用簽名時。在圖18中所示之實例中，Hap I及Hap II在基因座1及基因座3處共用簽名。 因此，可使用單倍型解卷積測定移植器官之貢獻。單倍型對移植器官之貢獻的增加將適用於指示器官對血漿DNA增加的貢獻。在各種實施例中，基線水準可測定自不具有排斥反應之移植受體群體或具有排斥反應之移植受體群體。當使用具有排斥反應之受體時，基線水準可測定為低於來自具有排斥反應之移植受體群體之彼等。 如上所述，供體可具有兩個一致單倍型或受體亦可具有兩個一致單倍型。此外，供體及受體可共用單倍型。只要供體或受體具有獨特單倍型，可測定來自供體組織之游離之DNA分子之百分比的變化。在前者中，將在看見血漿(或其他樣品)中之供體獨特性單倍型之貢獻增加時偵測到排斥反應。在後者中，將在看見血漿中之受體獨特性單倍型之貢獻減少時偵測到排斥反應。 因此，一些實施例可使用存在於生物體之正常細胞中且不存在於可在混合物中之異常細胞中的第一單倍型。此將對應於上文後一實例，當受體具有獨特單倍型時。另一實例為當患者具有相比於腫瘤之健康細胞中之獨特單倍型時(例如先前發現於生物體中)。在此實施例中，當第一分離值小於臨限值時，第一組織類型可測定為具有疾病病況。 在一些實施例中，若移植器官偵測為遭排斥，則可提供處理。舉例而言，可提供為抗排斥反應藥物之劑量的變化。作為另一實例，可獲得新器官，且可進行手術以移除舊移植的器官且放入新移植的器官。B . 肝細胞癌 ( HCC ) 作為測定癌症簽名或偏差之源組織(或監測已知存在或已存在之腫瘤)之實例，吾人分析HCC患者之血漿。患者之腫瘤及血細胞經定序以鑑別癌症特異性單核苷酸突變。攜有癌症特異性突變及至少一個CpG位點之經定序DNA片段用於下游甲基化解卷積分析。總共11,968個片段用於解卷積分析。除來自正常組織器官之甲基化圖譜以外，吾人亦包括HCC組織之甲基化圖譜作為候選源組織。 在另一實施例中，更多類型之腫瘤組織可視為用於突變之候選組織。在一個實施例中，常見癌症，例如(但不限於)結腸直腸癌、肺癌、乳癌、胰臟癌、前列腺癌、膀胱癌、子宮頸癌及卵巢癌之甲基化圖譜可包括為候選組織。在另一實施例中，僅最可能的特異於患者之癌症可包括於分析中。舉例而言，在女性患者中，考慮乳癌、卵巢癌、結腸直腸癌及子宮頸癌。在另一實施例中，在候選組織之選擇中考慮種族本源及年齡。 表7顯示攜有癌症相關突變之血漿DNA片段之甲基化解卷積。解卷積分析精確測定攜有癌症相關突變之DNA片段主要衍生自肝癌組織。 組織 貢獻 ( % )   肝臟 0.0 肺 0.0 結腸 0.0 小腸 0.0 胰臟 0.0 腎上腺 0.0 食道 0.0 脂肪組織 0.0 心臟 0.0 大腦 0.0 T細胞 0.0 B細胞 0.0 中性粒細胞 4.6 肝癌 95.4 胎盤 0.0 表 7 . 使用癌症突變之HCC患者之百分比貢獻。 在一些實施例中，腫瘤可起初藉由偵測拷貝數變異鑑別，例如如美國專利第8,741,811號及第9,121,069號中所述。可基於各種腫瘤中先前鑑別之拷貝數變異之模式測定特定源組織，例如如美國專利申請案14/994,053中所述。一旦已鑑別腫瘤，可進行治療，例如藉由手術、放射線療法或化學療法。無論如何，可在測定源組織之後獲得生檢。癌症特異性點突變可測定自生檢或測定自血漿中之DNA片段(例如如美國專利公開案2014/0100121中所述，或與拷貝數變異相關之其他混合物。 在治療後，關鍵變化將為基因組偏差，包括拷貝數變異及點突變之消失。當此等偏差消失時，受影響區域中之點突變之基因組簽名之分析將經由甲基化解卷積分析得到組織貢獻之變化。若腫瘤在將來恢復，則將再次可見組織組成中之癌症相關變化(如使用甲基化解卷積分析測定)。舉例而言，百分比貢獻可相比於參考百分比貢獻，且若偵測到變化，則可提供新治療時程。 在各種實施例中，癌症特異性突變可僅在一種單倍型上或在兩種單倍型上，例如以與上文供體實例類似之方式。因此，若存在不同簽名，則如同供體，可分別對於Hap I及Hap II中之每一者測定百分比貢獻；此類不同簽名可存在於不同基因座處。在其他實施例中，可對於兩種單倍型測定單一百分比貢獻，例如當其共用簽名時。C . 印跡 在另一實施例中，單倍型解卷積分析可應用於顯示組織特異性印跡之基因組區域之分析。已顯示不同組織器官中之父體及母體遺傳等位基因之差異性甲基化為正常現象(Baran等人 Genome Res 2015;25:927-36)。單倍型解卷積將適用於監測展現組織特異性印跡之器官的貢獻。舉例而言，當父體及母體遺傳單倍型在肝臟中但不在其他組織中具有不同甲基化狀態時，可對父體及母體遺傳單倍型兩者進行甲基化解卷積。在一個實施例中，父體及母體甲基化模式均可包括為分析中之候選組織。D . 使用基因組簽名之方法 圖19為說明分析生物體之生物樣品以偵測第一組織類型是否具有與根據本發明之實施例之第一單倍型相關之疾病病況之方法1900的流程圖。生物樣品包括來自複數種組織類型(包括第一組織類型)之游離之DNA分子之混合物。至少部分使用電腦系統進行方法1900。 在步驟1910處，分析來自生物樣品之複數個游離之DNA分子。步驟1910可使用圖1之方法100之步驟140中所述之技術進行。舉例而言，可分析至少1,000個游離之DNA分子以測定游離之DNA分子所位於之位置，且甲基化程度可如下所述地量測。另外，分析游離之DNA分子以測定游離之DNA分子之各別等位基因。舉例而言，DNA分子之等位基因可測定序列讀數或雜交至DNA分子之特定探針。 在步驟1920處，鑑別一或多個基因座。各基因座在第一染色體區域之第一單倍型上具有第一等位基因。第一單倍型具有以下特性中之任一者：(1)不存在於生物體之健康細胞中，但可替代地來自腫瘤或移植組織(作為實例)；或(2)存在於生物體之正常細胞中且不存在於可在混合物中之異常細胞中。因此，第一單倍型具有基因組簽名。以此方式，健康(正常)細胞與異常細胞之間存在差異，進而允許實施例追蹤一者或另一者或兩者之百分比貢獻，以追蹤異常細胞之程度(例如百分比貢獻)。在特性(1)之情況下，第一單倍型與疾病病況，例如癌症或移植組織之排斥反應相關聯。因此，特定癌症可在該特定癌症之癌症基因組中具有第一單倍型。 可藉由獲得組織樣品(例如腫瘤或移植組織)及分析組織樣品之DNA分子以測定第一單倍型來鑑別第一單倍型上之一或多個基因座處之一或多種第一等位基因。此類組織樣品可獲自生檢，且方法1900可用於測試癌症是否已轉移至其他組織，或是否在術後復發。基因座中的每一者可為異常細胞中之雜合基因座或純合基因座。舉例而言，在圖18中，基因座1及基因座3在供體器官中純合。但最後，將對於所有基因座在血漿中觀測到超過一個等位基因，因為各基因座將對於健康細胞或對於異常細胞具有簽名。因此，將跨越組織類型存在兩種單倍型，但單一組織類型可在分析區域中僅具有一種單倍型。 在步驟1930處，鑑別第一組複數個游離之DNA分子。複數個游離之DNA分子中之每一者位於來自步驟1920之基因座中的任一者處且包括一個基因座處之對應第一等位基因，以使得游離之DNA分子可鑑別為對應於第一單倍型。第一組游離之DNA分子中之每一者亦包括N個基因組位點中之至少一者，其中基因組位點用於量測甲基化程度。N為整數，例如大於或等於2、3、4、5、10、20、50、100、200、500、1,000、2,000或5,000。 在步驟1940處，使用第一組複數個游離之DNA分子量測N個基因組位點之N個第一混合物甲基化程度。可對於N個基因組位點中之每一者量測一個第一混合物甲基化程度。步驟1940可以與圖1之方法100之步驟150類似之方式進行。在一些實施例中，DNA分子之甲基化程度之量測可使用甲基化感測定序結果，其亦可用於測定DNA分子之位置及各別等位基因。 在步驟1950處，使用N個第一甲基化程度測定混合物中之第一組織類型之第一百分比貢獻。在一些實施例中，步驟1950可經由圖1之方法100之步驟160及170進行。因此，可對於一組M種組織類型同時測定百分比貢獻。步驟1950可使用對於M種組織類型中之每一者測定之N個基因組位點之N個組織特異性甲基化程度，例如如同圖1之方法100之步驟120。 在步驟1960處，計算第一百分比貢獻與參考百分比貢獻之間的分離值。分離值之實例描述於本文中。參考百分比貢獻可使用來自就第一組織類型而言健康之生物體的樣品測定。對於移植實例，參考百分比貢獻可測定自移植之第一組織不遭排斥之生物體之生物樣品的一或多次量測。 在步驟1970處，分離值可相比於臨限值以確定第一組織類型是否具有疾病病況之分類。舉例而言，若第一單倍型與癌症相關聯，則可觀的第一百分比貢獻指示第一組織類型具有癌症，如可藉由超出臨限值之分離值量測(例如當參考百分比貢獻為零時)。第一百分比貢獻超過臨限值之量可指示癌症之某一程度。作為另一實例，第一單倍型可特異於移植組織，且相對於參考之高貢獻可指示生物體排斥移植組織。 在第一單倍型存在於生物體之正常細胞中且不存在於可在混合物中之異常細胞中的一個實施例中，當第一分離值小於臨限值時，第一組織類型可測定為具有疾病病況。疾病病況之實例為先兆子癇，其可與如胎盤之胎兒組織之病理學變化譜相關聯。舉例而言，在此類情況下，若第一單倍型特異於胎兒，例如父體遺傳單倍型，則其可在併發先兆子癇之孕婦之母體血漿中增加。 在一些實施例中，亦可使用用於患病組織，例如移植組織或腫瘤之第二單倍型。因此，可計算第二百分比貢獻且相比於參考百分比貢獻。因此，第二組複數個游離之DNA分子可各位於一或多個基因座中的任一者處，包括第一染色體區域之第二單倍型上之對應第二等位基因，且包括N個基因組位點中之至少一者。第二單倍型將具有僅來自健康細胞或異常細胞之相同特性。 可測試複數種組織類型(例如使用圖1之方法100)，以測定第一單倍型之源組織，例如當其與癌症相關聯時。因此，可使用N個第一甲基化程度測定混合物中之其他組織類型之百分比貢獻，且對應百分比貢獻與各別參考百分比貢獻之間的對應分離值可相比於臨限值以確定其他組織類型中之每一者是否具有特定癌症之分類。不同組織可具有不同參考百分比貢獻。VIII. 鑑別癌症之 CNA 之源組織 在一些實施例中，腫瘤之來源可並非已知。因此，可能難以鑑別腫瘤中之點突變，如可用於圖19之方法1900或本文所述之其他方法。另外，腫瘤可不具有大量點突變，但可具有展現擴增及缺失(拷貝數變異之實例)之染色體區域。 為了解決此問題，實施例可使用複本數分析鑑別展現拷貝數變異(CNA)之區域。通常，CNA僅出現於區域之一種單倍型上。由於僅一種單倍型具有擴增或缺失，將在腫瘤駐留之組織類型之百分比貢獻之間存在相對較大差異。 CNA分析可以多種方式進行，例如如美國專利第8,741,811號及第9,121,069號中所述。舉例而言，人類基因組(或其他類型的生物體之基因組)可分割成大致3,000個非重疊1 Mb分區。可測定映射至各1 Mb分區之讀數數目。在對GC偏差進行校正(Chen EZ等人 (2011)PLoS One 6(7):e21791)之後，可計算各分區之序列讀數密度。對於各分區，測試案例之序列讀數密度可相比於參考對照個體之值。複本數增減可分別定義為高於及低於對照之平均值3個標準差。因此，可基於位於第一染色體區域中之游離之DNA分子之第一量將第一染色體區域鑑別為展現拷貝數變異。 為了測定血漿中之拷貝數變異之組織來源，可使用位於展現血漿中之此類偏差之基因組區域內之甲基化標記物進行血漿DNA組織映射。在關於癌症患者之以下實例中，僅在偏差影響至少30 Mb之連續染色體區域之情況下進行血漿DNA拷貝數變異之映射以使得足夠數目之甲基化標記物可用於映射。A . 鑑別具有拷貝數變異 ( CNA ) 之區域 在知情同意之情況下自香港威爾士王子醫院手術部(Department of Surgery, Prince of Wales Hospital, Hong Kong)募集患有HCC之62歲男性患者。在診斷時及切除腫瘤之後3個月在EDTA管中收集十毫升靜脈血液。血液樣品在3000 g下離心10分鐘以自血漿分離血細胞。血漿在30000 g下再離心10分鐘以移除其餘的細胞。 自血細胞提取之DNA用於遵循製造商說明書使用10x基因組學平台定相SNP以構築患者之單倍型。使用MagAttract HMW DNA套組(QIagen, Germany)自血液或組織樣品提取高分子量DNA。藉由基因組DNA分析ScreenTape在4200 TapeStation系統(Agilent, Germany)上驗證DNA之品質。DNA藉由dsDNA HS分析套組在Qubit 3.0螢光計(Thermo Fisher Scientific, Waltham, MA)上定量。使用GemCode系統及其相關試劑(10X Genomics, Pleasanton, CA)進行樣品索引及文庫製備(Zheng等人 Nat Biotechnol. 2016年3月;34:303-11)。簡言之，對於GEM反應輸入DNA之lng，其中個別DNA分子經分割以引入特定條碼且延長DNA。在GEM反應之後，根據製造商的建議製備定序文庫。文庫藉由使用KAPA文庫定量套組(KAPA Biosystems, Wilmington, MA)之qPCR定量。藉由98 bp、14 bp I5及8 bp I7索引讀數之雙末端定序在HiSeq 2500定序器(Illumina, San Diego, CA)上對標準化文庫定序。使用Long Ranger軟體程序組(10X Genomics)分析定序結果以使得所有雜合SNP經定相且測定患者之兩種單倍型。 血漿樣品使用Illumina定序17×之深度。拷貝數變異根據如先前所述之方法偵測於HCC患者之血漿中(Chan等人 Clin Chem. 2013;59:211-24)。 圖20顯示根據本發明之實施例在HCC患者之血漿中偵測之拷貝數變異之圖。內圓表示診斷時收集之血漿樣品之結果(預操作)且外圓表示在切除腫瘤之後3個月收集之血漿樣品之結果(後操作)。各點表示1 Mb區域。綠點、紅點及灰點分別表示複本數增加、複本數損失及無複本數變化的區域。拷貝數變異在診斷時偵測於血漿樣品中且此等變化在移除腫瘤之後消失。 在圖20中，兩個區域由於具有CNA而突出顯示。區域2010具有複本數增加，且區域2020具有複本數損失。可使用個體之任何組織樣品，且不僅僅腫瘤樣品來測定此等區域之單倍型。複本數之差異為驅動百分比貢獻之差異的因素，且差異應在具有腫瘤之組織類型中最大。B . 測定拷貝數變異之組織來源 吾人對於兩種單倍型獨立地進行甲基化解卷積分析。出於說明目的，兩種單倍型命名為Hap I及Hap II。覆蓋雜合SNP及至少一個CpG位點之血漿DNA分子用於此分析。在Hap I上攜有SNP等位基因之血漿DNA分子與在Hap II上攜有等位基因之彼等獨立地分析。CpG位點之甲基化狀態用於甲基化解卷積以將分子獨立地映射至Hap I及Hap II。因此，可測定對血漿DNA中之Hap I及Hap II之組織貢獻。 首先，吾人聚焦於具有擴增之區域。出於說明目的，吾人分析染色體lq(作為實例)上之擴增區域。    診斷時 腫瘤切除後 Hap I 34,119 11,131 Hap II 26,582 11,176 表11顯示來自兩種單倍型之序列讀數數目。在診斷時，映射至Hap I之讀數數目相比於映射至Hap II之讀數數目增加。此指示Hap I相對於Hap II擴增。此觀測結果與特定染色體在癌症而非擴增至相同程度之兩者同源染色體中複製之事實相容，其與拷貝數變異優先出現於一種單倍型之事實一致(Adey A.等人, Nature. 2013;500:207- 11；LaFramboise T.等人, PLoS Comput Biol. 2005;1(6):e65)。兩種單倍型之劑量差值在切除腫瘤之後消失。在診斷時與腫瘤切除之後獲取之血漿樣品之間的絕對序列讀數數目之差值係由於對於兩種血漿樣品產生之序列讀數之總數的差值。    診斷時 腫瘤切除後    Hap I Hap II 差值 Hap I Hap II 差值 肝臟 19.7 8.0 11.7 21.3 21.9 -0.6 肺 5.4 0 5.4 0 0 0 結腸 0 0 0 0 0 0 大腦 0 0 0 9.0 9.0 0 心臟 0 17.0 -17 3.0 2.5 -0.5 血細胞 74.9 75.0 0 66.7 66.6 0.1 總計 100 100 0 100 100 0 表12顯示針對在診斷時及腫瘤切除後之兩種單倍型的不同組織對血漿DNA之百分比貢獻。在診斷時，針對Hap I及Hap II的肝臟對血漿DNA之貢獻分別為19.7%及8.0%。不同類型的組織之間的最高差值為11.7%。此指示血漿中之Hap I與Hap II之間的劑量差值最可能貢獻自肝臟之貢獻。此另外指示染色體畸變之可能來源係來自肝臟，因為複本數變化最可能歸因於序列讀數計數分析中之Hap I之複製。在另一實施例中，Hap I及Hap II之貢獻之差值可經排位以指示不同組織為拷貝數變異之起源的相對似然性。 針對心臟之值為-17，其與藉由表11鑑別之拷貝數變異在相反方向上。因此，儘管針對心臟之絕對值大於針對肝臟之絕對值，相反符號將心臟折扣為腫瘤之源組織類型可行的候選物。由於所有器官之總貢獻為100%，肝臟貢獻之正差值導致其他組織具有負值。 類似地，此單倍型特異性甲基化解卷積亦可在具有複本數損失之區域上進行。出於說明目的，吾人在展現複本數損失之染色體1p上之區域上進行此分析。    診斷時 腫瘤切除後 Hap I 19,973 8,323 Hap II 12,383 7,724 表13顯示來自兩種單倍型之序列讀數數目。在診斷時，映射至Hap II之讀數數目相比於映射至Hap I之讀數數目減少。在腫瘤組織中，大部分具有染色體複本數損失之區域將僅涉及兩種染色體中的一者之缺失。因此，Hap II之劑量之相對減小與Hap II之缺失相容。在切除腫瘤之後消失的兩種單倍型之劑量差值指示腫瘤衍生之DNA的量已減少或自血漿消失。    診斷時 腫瘤切除後    Hap I Hap II 差值 (Hap I-Hap II) Hap I Hap II 差值 (Hap I-Hap II) 肝臟 13.3 5.5 7.8 10.2 13.2 -3 肺 0 0 0 4.1 0.5 3.6 結腸 3.8 0 3.8 8.6 17.5 -8.9 大腦 0 0 0 0 0 0 心臟 3.7 0 3.7 25.5 19.4 6.1 血細胞 79.2 94.5 -15.3 51.6 49.4 2.2 總計 100 100 0 100 100 0 表14顯示針對在診斷時及腫瘤切除後之兩種單倍型的不同組織對血漿DNA之百分比貢獻。在診斷時，針對Hap I及Hap II的肝臟對血漿DNA之貢獻分別為13.3%及5.5%。不同類型的組織之間的最高差值為7.8%。此指示血漿中之Hap I與Hap II之間的劑量差值最可能貢獻自肝臟之貢獻。此另外指示染色體畸變之可能來源係來自肝臟，因為複本數變化最可能歸因於序列讀數計數分析中之Hap II之缺失。在另一實施例中，Hap I及Hap II之貢獻之差值可經排位以指示不同組織為拷貝數變異之起源的相對似然性。C . 測定腫瘤之組織來源之方法 圖21為說明分析生物體之生物樣品以鑑別根據本發明之實施例之染色體畸變之起源之方法的流程圖。生物樣品包括來自包括第一組織類型之複數種組織類型之游離之DNA分子之混合物。 在步驟2110處，分析來自生物樣品之複數個游離之DNA分子。步驟2110可使用圖1之步驟1910及圖1之方法100之步驟140，以及描述類似特徵之其他步驟中所述之技術進行。 在步驟2115處，第一染色體區域基於位於第一染色體區域中之游離之DNA分子之第一量鑑別為展現生物體中之拷貝數變異。舉例而言，進行血漿DNA分析以鑑別展現拷貝數變異之區域。該畸變可對應於過表現或表現不足。在一些實施例中，基因組可分成分區(例如1 Mb分區)，且可測定來自特定分區之游離之DNA分子的量(例如藉由將序列讀數映射至參考基因組之該部分)。特定分區之量可經標準化(例如就分區之平均量而言)，且可鑑別過表現或表現不足。 可使用除計數向特定區域之DNA分子映射以外的其他技術。舉例而言，與第一染色體區域對準之DNA分子之尺寸分佈可用於偵測CNA。舉例而言，游離腫瘤DNA小於來自正常細胞之游離DNA。此尺寸差異可用於偵測該區域之兩種單倍型之間，或該區域與另一區域之間的尺寸分佈(例如平均尺寸或不同尺寸之DNA分子數目比)差異。 在步驟2120處，測定第一染色體區域中之生物體之第一單倍型及第二單倍型。兩種單倍型可測定為步驟2115之部分。兩種單倍型可使用相同游離混合物或自不同樣品，例如細胞樣品測定。 在步驟2130處，鑑別第一染色體區域之一或多個雜合基因座。各雜合基因座包括第一單倍型中之對應第一等位基因及第二單倍型中之對應第二等位基因。步驟2130可以與本文所述之方法之其他類似步驟類似之方式進行。 在步驟2140處，鑑別第一組複數個游離之DNA分子。第一組之各DNA分子位於一或多個雜合基因座中的任一者處，包括雜合基因座之對應第一等位基因，且包括N個基因組位點中之至少一者。N為大於或等於2之整數。步驟2140可以與本文所述之方法之其他類似步驟類似之方式進行。 在步驟2150處，使用第一組複數個游離之DNA分子量測N個基因組位點之N個第一混合物甲基化程度。步驟2150可以與本文所述之方法之其他類似步驟類似之方式進行。 在步驟2160處，鑑別第二組複數個游離之DNA分子。第二組之各DNA分子位於一或多個雜合基因座中的任一者處，包括雜合基因座之對應第二等位基因，且包括N個基因組位點中之至少一者。步驟2160可以與本文所述之方法之其他類似步驟類似之方式進行。 在一些實施例中，可測定第一組複數個游離之DNA分子中之游離之DNA分子之第一數目，且可測定第二組複數個游離之DNA分子中之游離之DNA分子之第二數目，例如如表11中所示。可測定何數目較高，進而向源組織提供關於預期分離值之資訊，例如何種單倍型應具有較高百分比貢獻。 第一組複數個游離之DNA分子可具有第一尺寸分佈，且第二組複數個游離之DNA分子可具有第二尺寸分佈。可對於各單倍型測定DNA分子之尺寸分佈之統計值，進而提供第一統計值及第二統計值。具有較小尺寸分佈之單倍型將預期具有比其他單倍型高的複本數，因為已知腫瘤游離DNA較小，如美國專利第8,741,811號中所述。尺寸分佈之統計值之實例為不同尺寸之DNA分子之數目比、平均尺寸或特定尺寸(例如低於尺寸閾值)之DNA分子之百分比。 在步驟2170處，使用第二組複數個游離之DNA分子量測N個基因組位點之N個第二混合物甲基化程度。步驟2170可以與本文所述之方法之其他類似步驟類似之方式進行。 步驟2180及2190可對於複數種M組織類型中之每一者進行。M種組織類型可包括經篩選且可已知參考甲基化程度之組織類型的預設清單。預設列表可包括最顯著可見癌症之組織。M為大於1之整數。 在步驟2180處，電腦系統使用N個第一甲基化程度測定混合物中之組織類型之對應第一百分比貢獻。電腦系統使用N個第二甲基化程度測定混合物中之組織類型之對應第二百分比貢獻。步驟2180可以與本文所述之方法之其他類似步驟類似之方式進行。 在步驟2190處，計算對應第一百分比貢獻與對應第二百分比貢獻之間的對應分離值。可使用各種分離值，例如如本文所述。 在步驟2195處，第一組織類型基於在對應分離值中具有最大值之第一組織類型之第一分離值鑑別為拷貝數變異之起源。測定可能需要最高分離值充分高於第二最高分離值。舉例而言，可能需要差值至少為臨限值，例如1%、2%、3%、4%、5%、6%或7%。在一個實施方案中，第一分離值與次最高分離值之間的差值可相比於臨限值以確定第一組織類型為拷貝數變異之起源之可能性程度的分類。因此，即使該差值不高於臨限值，可提供概率或其他分類。舉例而言，可使用0至臨限值之線性關係，其中一旦該差值等於臨限值，概率為100%。 取決於如何測定分離值，最大值可為最大負數或最大正數。舉例而言，表14中之差值可使用Hap II-Hap I測定。可使用各單倍型上之DNA分子之分析測定最大值應為正值或負值，例如如表13中之計數或如上文所述之尺寸分析。在一些實施方案中，可始終測定分離值以使得預期最大正值，例如藉由自具有較高複本數之單倍型之百分比貢獻減去具有較低複本數之單倍型之百分比貢獻。 在鑑別來源之後，可進行使用成像模式，例如個體(整個個體或特定言之，候選器官)之電腦層析成像(CT)掃描或磁共振成像(MRI)之研究以證實或排除腫瘤於器官中之存在。若證實腫瘤存在，則可進行治療，例如手術(藉由手術刀或藉由輻射)或化學療法。IX . 電腦系統 本文中提及之任何電腦系統均可利用任何適合數目之子系統。此類子系統之實例顯示於圖22之電腦設備10中。在一些實施例中，電腦系統包括單一電腦設備，其中子系統可為電腦設備之組件。在其他實施例中，電腦系統可包括具有內部組件之多個電腦設備，其各自為一個子系統。電腦系統可包括桌上型及膝上型電腦、平板電腦、行動電話及其他行動裝置。 圖22中所示之子系統經由系統匯流排75互連。展示額外子系統，諸如印表機74、鍵盤78、儲存裝置79、耦接至顯示配接器82之監測器76等。耦合至輸入/輸出(I/O)控制器71之周邊裝置及I/O裝置可藉由任何數目之此項技術中已知之構件(諸如輸入/輸出(I/O)埠77 (例如，USB、FireWire® ))連接至電腦系統。舉例而言，I/O埠77或外部介面81 (例如，乙太網路、Wi-Fi等)可用於將電腦系統10連接至廣域網路(諸如，網際網路、鼠標輸入裝置或掃描儀)。經由系統匯流排75實現之互連允許中央處理器73與各子系統通信及控制來自系統記憶體72或儲存裝置79 (例如固接磁碟，諸如硬碟機，或光碟)之複數個指令之執行，以及子系統之間的資訊交換。系統記憶體72及/或儲存裝置79可體現電腦可讀媒體。另一子系統為資料採集裝置85，諸如，攝影機、麥克風、加速計及其類似物。本文所提及之任何資料可自一個組件輸出至另一個組件且可輸出至使用者。 電腦系統可包括例如藉由外部接口81或藉由內部接口連接在一起的複數個相同組件或子系統。在一些實施例中，電腦系統、子系統或設備可經網路連通。在該等情況下，可將一個電腦視為用戶端且另一個電腦視為伺服器，其中每一者可為同一電腦系統之一部分。用戶端及伺服器各自可包括多個系統、子系統或組件。 實施例之態樣可使用硬體(例如特殊應用積體電路或場可程式化閘陣列)、以邏輯控制形式實施及/或使用電腦軟體、使用普通可程式化處理器、以模組化或整合方式實施。如本文中所使用，處理器包括位於同一積體晶片上之單核心處理器、多核心處理器，或位於單一電路板上或網路化之多個處理單元。基於本文所提供之揭示內容及教示，一般熟習此項技術者將知道及瞭解使用硬體及硬體與軟體之組合來實施本發明之實施例的其他方式及/或方法。 描述於本申請案中之任何軟體組件或功能可作為待由處理器執行的使用任何適合之電腦語言(諸如Java、C、C++、C#、Objective-C、Swift)或腳本語言(諸如Perl或Python)的軟體程式碼，使用例如習知或目標定向技術來執行。軟體程式碼可以一系列指令或命令形式儲存於電腦可讀媒體上以用於儲存及/或傳輸。適合的非暫時性電腦可讀媒體可包括隨機存取記憶體(RAM)、唯讀記憶體(ROM)、磁性媒體(諸如硬碟機或軟碟機)，或光學媒體，諸如光盤(CD)或DVD (數位化通用光碟)、快閃記憶體，及其類似物。電腦可讀媒體可為該等儲存或傳輸裝置之任何組合。 該等程序亦可使用適用於經由有線、光學及/或符合多種協定之無線網路(包括網際網路)傳輸的載波信號來編碼及傳輸。因而，電腦可讀媒體可使用經由此類程式編碼的資料信號建立。以程式碼編碼之電腦可讀媒體可與相容裝置一起封裝或與其他裝置分開單獨提供(例如藉助於網際網路下載)。任何此等電腦可讀媒體可存在於單一電腦產品(例如，硬碟機、CD或整個電腦系統)上或其內部，且可存在於系統或網路內之不同電腦產品上或其內部。電腦系統可包括用於向使用者提供本文所提及之任何結果的監測器、印表機、或其他適合之顯示器。 本文所描述之任何方法可完全或部分地使用電腦系統來進行，該電腦系統包括一或多個處理器，該等處理器可經組態以進行該等步驟。因此，實施例可針對經組態以執行本文所描述之任何方法之步驟的電腦系統，潛在地使用不同組件執行各別步驟或各別步驟組。儘管本文中方法之步驟以經編號之步驟呈現，但其可同時或以不同次序執行。另外，此等步驟之部分可與其他方法之其他步驟之部分一起使用。另外，步驟之全部或部分可視情況選用。另外，任何方法中之任何步驟可使用用於執行此等步驟的模組、單元、電路或其他構件來執行。 可在不脫離本發明之實施例的精神及範疇的情況下以任何適合之方式組合特定實施例之特定細節。然而，本發明之其他實施例可針對於與各個別態樣或此等個別態樣之特定組合相關的特定實施例。 已出於說明及描述之目的呈現本發明之實例實施例的上述描述。其並不意欲為窮盡性的或將本發明限制於所描述之精確形式，且鑒於以上教示，許多修改及變化為可能的。 除非特別指示相反，否則「一(a/an)」或「該(the)」之敍述意欲意謂「一或多個(種)」。除非明確指示相反，否則「或」之使用欲意謂「包括或」而並非「互斥或」。提及「第一」組件不一定需要提供第二組件。此外，除非有明確陳述，否則提及「第一」或「第二」組件不會將所提及組件限於特定位置。 本文所提及之所有專利、專利申請案、公開案及描述均出於所有目的以全文引用之方式併入。不容許任一者為先前技術。[Cross-reference to related applications] This application claims priority to U.S. Provisional Application No. 62/194,702, filed July 20, 2015, entitled "Methylation Pattern Analysis Of Haplotypes In Tissues In A DNA Mixture," the entire contents of which are incorporated by reference for all purposes manner is incorporated into this article. the term A "methylome" provides a measure of the amount of DNA methylation at multiple sites or loci in the genome. A methylome can correspond to the entire genome, a large portion of a genome, or a relatively small portion of a genome. Examples of methylomes of interest are those that can contribute DNA to body fluids (eg, plasma, serum, sweat, saliva, urine, genital secretions, semen, fecal fluid, diarrhea fluid, cerebrospinal fluid, gastrointestinal secretions, Ascites fluid, pleural fluid, intraocular fluid, fluid from hydrocysts (e.g. testis), fluid from cysts, pancreatic secretions, intestinal secretions, sputum, tears, aspirated fluids from breast and thyroid, etc.) The methylome of organs in the brain (such as the methylome of brain cells, bone, lung, heart, muscle and kidney, etc.). The organ may be a transplanted organ. The methylome of the fetus is another example. A "plasma methylome" is a methylome determined from plasma or serum of animals (eg, humans). The plasma methylome is an example of the free methylome, since plasma and serum include free DNA. Plasma methylomes are also examples of mixed methylomes, as either embryonic/maternal methylomes or tumor/patient methylomes or derived from different tissues or organs or donors in the background or in organ transplantation/ A mixture of DNA from the receptor methylome. A "site" (also referred to as a "genomic site") corresponds to a single site, which can be a single base position or a group of related base positions, such as a CpG site or a larger group of related base positions. A "locus" can correspond to a region that includes multiple loci. A locus can include only one site, which would make the locus equivalent to one site in this context. The "methylation index" for each genomic locus (eg, a CpG site) can refer to the DNA fragments (eg, as determined from sequence reads or probes) that exhibit methylation at that locus relative to the total number of reads covering the locus ratio. A "read" can correspond to information obtained from a DNA fragment (eg, the methylation status of a site). Reads can be obtained using reagents (eg, primers or probes) that preferentially hybridize to DNA fragments in a particular methylation state. Typically, such agents are used in methods that modify or recognize DNA molecules differently depending on their methylation status, such as bisulfite conversion, or methylation-sensitive restriction enzymes, or methylation-binding proteins , or anti-methylcytosine antibody, or single-molecule sequencing technology that recognizes methylcytosine and hydroxymethylcytosine and administered after treatment. The "methylation density" of a region may refer to the number of reads at sites within the region showing methylation divided by the total number of reads at sites in the coverage region. Sites may have specific characteristics, such as CpG sites. Thus, the "CpG methylation density" of a region can refer to the number of reads showing CpG methylation divided by the number of CpG sites in the covered region (eg, a specific CpG site, CpG island, or CpG sites within a larger region) Total number of readings. For example, the methylation density per 100 kb bin in the human genome can be determined from the total number of unconverted cytosines (which correspond to methylated cytosines) at CpG sites after bisulfite treatment as Proportion of all CpG sites covered by sequence reads mapped to a 100 kb region. This analysis can also be performed for other partition sizes such as 500 bp, 5 kb, 10 kb, 50 kb or 1 Mb, etc. A region can be the entire genome or a chromosome or part of a chromosome (eg, a chromosome arm). When the region includes only CpG sites, the methylation index of the CpG sites is the same as the methylation density of the region. "Ratio of methylated cytosines" may refer to the number of cytosine sites "C" that exhibit methylation (eg, unconverted after bisulfite conversion) compared to the total number of cytosine residues analyzed, That is, cytosines in the region other than the CpG background are included. Methylation index, methylation density and ratio of methylated cytosines are examples of "degree of methylation". In addition to bisulfite conversion, other methods known to those skilled in the art can be used to interrogate the methylation status of DNA molecules, including, but not limited to, enzymes sensitive to methylation status (eg, methylation-sensitive restriction enzymes), methylation-binding proteins, single-molecule sequencing using platforms sensitive to methylation status (e.g., nanopore sequencing (Schreiber et al. Proc Natl Acad Sci 2013; 110: 18910-18915) and by Pacific Biosciences Single Molecule Real Time Analysis (Flusberg et al. Nat Methods 2010; 7: 461-465)). A "methylation profile" (also known as methylation status) includes information related to DNA methylation of a region. Information related to DNA methylation may include, but is not limited to, methylation index of CpG sites, methylation density of CpG sites in a region, distribution of CpG sites in adjacent regions, containing more than one CpG site The pattern or extent of methylation and non-CpG methylation of each individual CpG site within the region of the dots. Most methylation profiles of the genome can be considered equivalent to the methylome. "DNA methylation" in mammalian genomes generally refers to the addition of methyl groups to the 5' carbon of cytosine residues in CpG dinucleotides (ie, 5-methylcytosine). DNA methylation can occur in cytosine in other instances such as CHG and CHH, where H is adenine, cytosine or thymine. Cytosine methylation can also be in the form of 5-hydroxymethylcytosine. Also reported non-cytosine methylation, such as N6 - Methyladenine. "Methylation sensing sequencing" refers to any sequencing method that allows us to determine the methylation status of DNA molecules during the sequencing method, including but not limited to bisulfite sequencing, or pre-sequencing methyl Methylation-sensitive restriction enzyme digestion, immunoprecipitation using anti-methylcytosine antibodies or methylation-binding proteins, or single-molecule sequencing that elucidates methylation status. A "tissue" corresponds to a group of cells that are collectively classified as a functional unit. More than one type of cell can be found in a single tissue. Different types of tissues may be composed of different types of cells (eg, liver cells, alveolar cells, or blood cells), but may also correspond to tissues from different organisms (mother and fetus) or to healthy cells and tumor cells. A "reference tissue" corresponds to the tissue used to determine the degree of tissue-specific methylation. Multiple samples of the same tissue type from different individuals can be used to determine the degree of tissue-specific methylation of that tissue type. "Biological sample" means obtained from an individual (eg, a human, such as a pregnant woman, an individual with or suspected of having cancer, a recipient of an organ transplant or suspected of having an organ involved (eg, the heart in myocardial infarction, or in a stroke) of the brain, or the hematopoietic system in anemia) and any sample that contains one or more nucleic acid molecules of interest. The biological sample can be a body fluid, such as blood, plasma, serum, urine, vaginal fluid, fluid from a hydrocele (eg, testis), or vaginal irrigation fluid, pleural fluid, ascites fluid, cerebrospinal fluid, saliva, sweat, tears, Sputum, bronchoalveolar lavage fluid, etc. Fecal samples may also be used. In various embodiments, a substantial portion of the DNA in a biological sample (eg, a plasma sample obtained via a centrifugation protocol) in which cell-free DNA has been enriched may be free (as opposed to cells), eg, greater than 50%, 60%, 70%, 80% , 90%, 95% or 99%. The centrifugation protocol may include 3,000 g x 10 minutes, obtaining a fluid fraction and centrifugation at 30,000 g for an additional 10 minutes to remove residual cells. The term "cancer extent" can refer to the presence or absence of cancer (ie, presence or absence), cancer stage, tumor size, presence or absence of metastases, total body tumor burden, and/or other measures of cancer severity (eg, cancer recurrence). The degree of cancer can be a number or other signs, such as symbols, alphabet letters and colors. The degree can be zero. The extent of cancer also includes precancerous or precancerous conditions (states) associated with the mutation or mutations. Cancer levels can be used in various ways. For example, screening can check for the presence of cancer in someone who is known not to have had cancer before. An assessment can investigate someone who has been diagnosed with cancer to monitor the progression of the cancer over time, to study the effectiveness of a treatment, or to determine a prognosis. In one embodiment, prognosis can be expressed in terms of the probability that a patient will die from the cancer or the probability that the cancer will progress or the cancer will metastasize after a certain period or time. Detecting can mean "screening" or it can mean checking whether someone who has features suggestive of cancer (such as symptoms or other positive tests) has cancer. The term "sequence imbalance" of a chromosomal region can refer to any significant deviation in the amount of free DNA molecules from a chromosomal region relative to the expected value in the case of a healthy organism. For example, a chromosomal region may exhibit an amplification or deletion in a certain tissue, resulting in an imbalance in the sequence of the chromosomal region in a DNA mixture containing DNA from a tissue mixed with DNA from other tissues. For example, the expected value can be obtained from another sample or from another chromosomal region that is assumed to be normal (eg, an amount representing two copies of a diploid organism). A chromosomal region may consist of multiple disjoint subregions. The "type" of a genomic locus (marker) corresponds to a specific property of the locus across tissue types. The description mainly refers to the type I locus and the type II locus, the characteristics of which are provided in detail below. A locus of a given type can have a specific statistical variation in the degree of methylation across tissue types. A "class" of a genomic locus (marker) corresponds to a specific variation in the degree of methylation of the locus across different individuals of the same tissue type. A set of loci (markers) can consist of any number of loci of various types and/or classes. Thus, a set of loci corresponds to a locus selected for a particular measurement and does not imply any particular properties of the loci in the set. A "separated value" corresponds to the difference or ratio involving two values, eg, two percent contributions or two degrees of methylation. Separation values can be simple differences or ratios. Separate values may include other coefficients, such as multiplication coefficients. As other examples, the difference or ratio of functions of the equivalents, such as the difference or ratio of the natural logarithms (ln) of two values, may be used. Separate values can include differences and ratios. As used herein, the term "Classification ” means any number or other character associated with a particular characteristic of the sample. For example, the symbol "+" (or the word "positive") may indicate that the sample is classified as having a deletion or amplification. Classifications can be binary (eg, positive or negative) or have more classification levels (eg, a scale of 1 to 10 or 0 to 1). the term"threshold "and"Threshold value ” means a predetermined number used for operation. Threshold values may be above or below the value applicable for a particular classification. Either of these terms may be used in any of these contexts. Methylation differences between tissue types (eg, embryonic tissue, liver, etc.) in DNA mixtures (eg, plasma) can be used to distinguish the characteristics of tissue haplotypes for specific tissue types. For example, the degree of methylation of two maternal haplotypes in the plasma of a pregnant woman can be used to determine which haplotypes are inherited from the mother to the fetus. As another example, the degree of methylation of two haplotypes in embryonic tissue can be used to detect sequence imbalances (eg, aneuploidy) in the fetus. Other tissue types can also be analyzed, eg, to detect disease conditions in a particular tissue type. The tissue type from which the copy number variation originates can also be determined. Some embodiments can determine the percentage of cell-free DNA in plasma (or other DNA mixture) from various tissue types using the known methylation levels of certain genomic loci in a particular tissue type. For example, the degree of methylation at genomic loci can be measured for liver samples, and these tissue-specific methylation levels can be used to determine how much cell-free DNA in a mixture comes from liver. The degree of methylation of tissue types that provide a significant contribution to the DNA mixture can be measured so that the dominance of the cell-free DNA mixture (eg, greater than 90%, 95%, or 99%) can be calculated. Such other samples may include, but are not limited to, some or all of the following: lung, colon, small intestine, pancreas, adrenal gland, esophagus, adipose tissue, heart, and brain. Deconvolution methods can be used to determine the percent contribution (eg, percent) of each of the tissue types of known tissue-specific methylation levels. In some embodiments, a system of linear equations can be generated from known tissue-specific methylation levels and a mixture of methylation levels at a given genomic locus, and the best approximation of the measured mixture methylation can be determined (eg, using least squares) The percentage contribution to the degree of transformation. Once the percent contribution is determined, the percent contribution can be used for various purposes. For example, the difference in percentage contribution of embryonic tissue can be used to determine which haplotype is inherited from the parent. Alleles at one or more heterozygous loci can be determined for each of the two parental haplotypes. Episomal DNA at one or more heterozygous loci can be used to determine two percent contributions: one for each haplotype. For example, an episomal DNA molecule with an allele of a first haplotype can be used to determine a first percent contribution, and an episomal DNA molecule with an allele of a second haplotype can be used to determine a second percent contribution. Inherited haplotypes will correspond to higher percentage contributions of embryonic tissue. In addition, the inherited haplotype will have a lower degree of methylation due to the general hypomethylation of embryonic cell-free DNA. The degree of methylation of the two haplotypes can be compared, and the haplotype with the lower degree of methylation can be identified as the inherited haplotype. As another example, a sequence imbalance can be detected in a chromosomal region of interest in a fetus. The target percent contribution of embryonic tissue types in the mixture of the first haplotype in the target chromosomal region can be determined. Similarly, the reference percent contribution of embryonic tissue types for reference chromosomal regions can be determined. The separation value between the two contributions can be compared to a threshold value to determine whether the fetus has a sequence imbalance (eg, aneuploidy). As another example, the first haplotype may have a signature specific for healthy cells or abnormal cells. The separation value between the percent contribution determined for the first haplotype and the reference percent contribution can be compared to a threshold value to determine whether the first tissue type has classification of a disease condition. As an example, the first haplotype may be in the transplanted organ or tumor, or only in healthy cells and no longer in the transplanted organ or tumor. The disease condition can be whether the transplanted organ has been rejected, or whether the tumor has increased in size or has metastasized (eg, after surgery that does not remove all the tumor). As another example, tissue sources of copy number variation can be determined using methylation deconvolution. The first chromosomal region can be identified as exhibiting copy number variation. For each of the M tissue types, the corresponding separation value between the percent contributions of the two haplotypes in the first chromosomal region can be determined. The tissue type with the highest separation value can be identified as the source tissue. Methylation deconvolution is described first, and then the selection of methylation markers and the accuracy of methylation deconvolution are described. The use of the percent contribution for determining portions of embryonic genomes is then described. I. Composition of DNA Mixtures Deconvoluted According to Methylation Different tissue types can have different degrees of methylation for genomic loci. These differences can be used to determine the percent contribution of DNA from various tissue types in the mixture. Thus, the composition of DNA mixtures can be determined by tissue-specific methylation pattern analysis. The following examples discuss methylation densities, although other degrees of methylation can be used.A . single genomic locus The principle of methylation deconvolution can be illustrated by determining the composition of DNA mixtures from an organism using a single methylated genomic locus (methylation marker). It is assumed that tissue A is fully methylated for the genomic locus, ie, a methylation density (MD) of 100% and tissue B is completely unmethylated, ie, a MD of 0%. In this example, methylation density refers to the percentage of cytosine residues in the context of CpG dinucleotides that are methylated in the region of interest. If DNA mixture C consists of tissue A and tissue B and the total methylation density of DNA mixture C is 60%, we can infer the proportional contribution of tissue A and B to DNA mixture C according to the following formula: MDC = MDA ×α + MDB × b, inMD A ,MD B ,MD C represent the MD of tissue A, tissue B, and DNA mixture C, respectively; and a and b are the proportional contributions of tissue A and B to DNA mixture C. In this particular example, tissues A and B are assumed to be the only two components of the DNA mixture. Therefore, a+b=100%. It was thus calculated that tissues A and B contributed 60% and 40%, respectively, to the DNA mixture. Methylation densities in tissue A and tissue B can be obtained from samples of the organism or from samples of other organisms of the same type (eg, other humans, potentially the same subpopulation). If samples from other organisms are used, statistical analysis (eg, mean, median, geometric mean) of methylation densities for samples from tissue A can be used to obtain methylation densitiesMD A , and forMD B The situation is similar. Genomic loci can be selected to have minimal inter-individual variation, eg, less than a specified absolute amount of variation or within the lowest portion of the genomic locus tested. For example, for the lowest fraction, an embodiment may select only the genomic loci that have the lowest 10% of variation in a population of genomic loci tested. Other organisms can be obtained from healthy individuals, as well as individuals with specific physiological conditions (eg, pregnant women, or people of different ages or genders), which can correspond to specific subpopulations of organisms comprising the current test. Other organisms of the subpopulation may also have other pathological conditions (eg, patients with hepatitis or diabetes, etc.). Such subsets may have altered tissue-specific methylation patterns for various tissues. In addition to using methylation patterns of normal tissues, methylation patterns of tissues in such disease conditions can be used for deconvolution analysis. This deconvolution analysis can be more accurate when testing organisms from such subpopulations with their conditions. For example, a cirrhotic liver or a fibrotic kidney may have different methylation patterns compared to normal liver and normal kidney, respectively. Therefore, if cirrhotic patients are screened for other diseases, it may be more precise to include cirrhotic liver as one of the candidates for contributing DNA to plasma DNA, along with healthy tissue of other tissue types.B . Multiple genomic loci When there are more potential candidate tissues, more genomic loci (eg, 10 or more) can be used to determine the composition of the DNA mixture. The accuracy of the assessment of the ratiometric composition of DNA mixtures depends on a number of factors, including the number of genomic loci, the specificity of the genomic locus (also referred to as "locus") for a particular tissue, and across different candidate tissues and across the reference tissue used for the assay Site variability across individuals in the degree of specificity. Site-to-tissue specificity refers to differences in methylation density at genomic loci in a particular tissue between other tissue types. The greater the difference between their methylation densities, the more specific the site will be to a particular tissue. For example, if a site is fully methylated (methylation density=100%) in the liver and fully unmethylated (methylation density=0%) in all other tissues, then this site will highly specific. However, site variability across different tissues can be reflected by, for example, but not limited to, the range or standard deviation of methylation densities at sites in different types of tissues. Larger ranges or higher standard deviations will allow mathematically more precise and precise determination of the relative contributions of different organs to the DNA mixture. The effect of these factors on the accuracy of estimating the proportional contribution of candidate tissues to the DNA mixture is described in subsequent sections of this application. Here, we use mathematical equations to illustrate the derivation of the proportional contributions of different organs to the DNA mixture. The mathematical relationship between the methylation densities of different sites in a DNA mixture and the methylation densities of corresponding sites in different tissues can be expressed as:
Figure 02_image005
, in
Figure 02_image007
represents the methylation density of site i in the DNA mixture;p k represents the proportional contribution of tissue k to the DNA mixture;MD ik represents the methylation density of site i in tissue k. When the number of sites is the same as or greater than the number of organs, individualp k value. Tissue-specific methylation densities can be obtained from other individuals, and sites can be selected to have minimal inter-individual variability, as described above. Additional criteria can be included in the algorithm to improve accuracy. For example, the aggregate contribution of all organizations can be limited to 100%, i.e.
Figure 02_image009
. In addition, the contributions of all organs may need to be non-negative:
Figure 02_image011
Due to biological variation, the observed overall methylation pattern may not be identical to the inferred methylation pattern of self-organized methylation. In such cases, mathematical analysis would be required to determine the most likely proportional contribution of individual tissues. In this regard, the difference between the observed methylation pattern in DNA and the self-organized inferred methylation pattern is indicated by W.
Figure 02_image013
where O is the methylation pattern observed for the DNA mixture andM k is the methylation pattern of individual tissue k.p k Contribute to the ratio of tissue k to the DNA mixture. eachp k The most likely value of W can be determined by minimizing W, which is the difference between the observed and inferred methylation patterns. This equation can be solved using mathematical algorithms such as, but not limited to, by using quadratic programming, linear/non-linear regression, expectation maximization (EM) algorithms, maximum likelihood algorithms, maximum a posteriori evaluation, and least squares methods.C . methylation deconvolution method As described above, biological samples comprising a mixture of free DNA molecules from an organism can be analyzed to determine the composition of the mixture, in particular the contribution from different tissue types. For example, the percent contribution of free DNA molecules from the liver can be determined. These measurements of percent contribution in biological samples can be used to perform other measurements of biological samples, such as identifying where tumors are located, as described in a later section. 1 is a flowchart illustrating a method 100 of analyzing a DNA mixture of free DNA molecules to determine the percent contribution from various tissue types from methylation levels according to embodiments of the present invention. Biological samples include a mixture of free DNA molecules from M tissue types. The biological sample can be any of various examples, eg, as mentioned herein. The number M of tissue types is greater than 2. In various embodiments, M can be 3, 7, 10, 20, or greater than 20, or any number in between. The method 100 can be performed at least in part using a computer system, as can other methods described herein. At step 110, N genomic loci are identified for analysis. The N genomic loci can have various properties, eg, as described in more detail in Section II, which describes Type I and Type II genomic loci. As an example, the N genomic loci may include only type I or type II loci, or a combination of the two. Genomic loci can be identified based on analysis of one or more other samples, eg, based on data obtained from a database of methylation levels measured in various individuals. Specific genomic loci can be selected to provide the desired degree of precision. For example, loci with at least one threshold variability can be used, as opposed to using only loci specific for one tissue type. A first set (eg, 10) of genomic loci can be selected such that each has a coefficient of variation of methylation degrees of at least 0.15 across M tissue types and such that each has M tissues exceeding 0.1 for one or more other samples The difference between the maximum and minimum methylation levels of the type. This first set of genomic loci may not have a specific methylation signature for a specific tissue type, eg, be methylated only or predominantly in a specific tissue type. This first group is called type II sites. These genomic loci can be used in combination with genomic loci with specific signatures called Type I loci. The use of Type II sites ensures that the full space of methylation levels across tissue types is spanned by genomic sites, thereby providing increased accuracy compared to Type I sites. Using only more type I sites provides redundant basis for the methylation space (i.e. more genomic sites with the same pattern as other sites), while adding methylation degrees has a variety of values across different tissues. Additional genomic loci added novel basis vectors to differentiate percent contributions via a system of linear equations. In some embodiments, at least 10 of the N genomic loci each have a coefficient of variation of methylation degree across M tissue types of at least 0.15. At least 10 genomic loci may also each have a difference between the maximum and minimum methylation levels of the M tissue types of more than 0.1. These methylation properties of loci can be measured for a sample or group of samples. A sample set can be directed to a subpopulation of organisms comprising the tested organisms of the present invention, eg, a subpopulation having particular characteristics in common with the present organisms. These other samples can be referred to as reference tissues, and different reference tissues from different samples can be used. At step 120, N tissue-specific methylation levels are obtained at N genomic loci for each of the M tissue types. N is greater than or equal to M so that tissue-specific methylation levels can be used for deconvolution to determine percent fraction. Tissue-specific methylation levels can form a matrix A of dimension N×M. Each row of matrix A may correspond to a methylation pattern for a particular tissue type, where the pattern has the degree of methylation at N genomic loci. In various embodiments, tissue-specific methylation patterns can be retrieved from public repositories or the aforementioned studies. In the examples herein, methylation data for neutrophils and B cells were downloaded from the Gene Expression Omnibus (Hodges et al. Mol Cell 2011;44:17-28). The methylation patterns of other tissues (hippocampus, liver, lung, pancreas, atrium, colon (including various parts thereof such as sigmoid, transverse, ascending, descending), adrenal gland, esophagus, small intestine, and CD4 T cells) are derived from RoadMap Epigenomics project (Ziller et al. Nature 2013; 500:477-81) downloaded. Methylation patterns for leukocyte, placenta, tumor, and plasma data were published from reports (Lun et al. Clin Chem. 2013;59:1583-94; Chan et al. Proc Natl Acad Sci USA. 2013;110:18761-8). These tissue-specific methylation patterns can be used to identify N genomic loci for use in deconvolution analysis. At step 130, a biological sample comprising a mixture of free DNA molecules from M tissue types is received. Biological samples can be obtained from patient organisms in a variety of ways. The manner in which such samples are obtained can be non-invasive or invasive. Examples of non-invasively obtained samples include certain types of fluids (eg, plasma or serum or urine) or feces. For example, plasma includes free DNA molecules from many organ tissues and is therefore suitable for the analysis of many organs via one sample. At step 140, free DNA molecules from the biological sample are analyzed to identify their corresponding locations in the organism's reference genome. For example, free DNA molecules can be sequenced to obtain sequence reads, and the sequence reads can be mapped (aligned) to a reference genome. If the organism is a human, the reference genome will be the reference human genome potentially from a particular subpopulation. As another example, free DNA molecules can be analyzed (eg, after PCR or other amplification) by different probes, where each probe corresponds to a genomic location that can cover both the heterozygote and one or more CpG sites, as described below. A statistically significant number of free DNA molecules can be analyzed to provide accurate deconvolution to determine the percent contribution from M tissue types. In some embodiments, at least 1,000 free DNA molecules are analyzed. In other embodiments, at least 10,000 or 50,000 or 100,000 or 500,000 or 1,000,000 or 5,000,000 or more than 5,000,000 free DNA molecules can be analyzed. The total number of molecules analyzed can depend on M and N, and the desired precision (precision). In various examples, the total number of cell-free DNA analyzed can be less than 500,000, 1 million, 2 million, 5 million, 10 million, 20 million, or 50 million. At step 150, the degree of methylation of the N mixture of N genomic loci is measured using the free DNA molecular weight each located at any of the N genomic loci of the reference genome. A DNA molecule can be identified as being located at a genomic locus or locus by one or more bases of the DNA molecule corresponding to one or more base positions of the genomic locus or locus. Thus, the sequence of the DNA molecule will cover one or more base positions of the genomic locus or locus. This information may be based on the location determination determined in step 140 . Such identification of DNA molecules at loci can be used in any analogous step of the methods described herein. The degree of methylation of the N mixture refers to the degree of methylation in the mixture of biological samples. For example, if a free DNA molecule from a mixture is located at one of N genomic loci, the methylation index of the molecule for that locus can be included in the overall methylation density for that locus. The degree of methylation of the mixture of N can form a methylation vector b of length N, where b corresponds to the observation from which the percent contribution of each corresponding tissue type can be determined. In one embodiment, whole genome bisulfite sequencing can be used to determine the degree of methylation at genomic loci in DNA mixtures. In other embodiments, methylation microarray analysis can be used, such as the Illumina HumanMethylation450 system, or by immunoprecipitation using methylation (eg, using anti-methylcytosine antibodies) or treatment with methylation-binding proteins, followed by by microarray analysis or DNA sequencing, or by treatment with methylation-sensitive restriction enzymes followed by microarray or DNA sequencing, or by use of methylation sensing sequencing, for example using single-molecule sequencing methods ( Determination of CpG positions, for example, by nanopore sequencing (Schreiber et al. Proc Natl Acad Sci 2013; 110: 18910-18915) or by Pacific Biosciences single-molecule real-time analysis (Flusberg et al. Nat Methods 2010; 7: 461-465) The degree of methylation at the point. Tissue-specific methylation levels can be measured in the same way. In other embodiments, other methods, such as, but not limited to, targeted bisulfite sequencing, methylation-specific PCR, non-bisulfite-based methylation sensing sequencing (eg, by single-molecule A sequencing platform (Powers et al. Efficient and accurate whole genome assembly and methylome profiling of E. coli. BMC Genomics. 2013; 14:675)) can be used to analyze the degree of methylation of plasma DNA from plasma DNA methylation deconvolution analysis . At step 160, the M value of the combined vector is determined. Each M value corresponds to the percent contribution of a particular tissue type in M to the DNA mixture. Given that matrix A consists of N×M tissue-specific methylation levels (that is, N tissue-specific methylation levels for each of M tissue types), the M-values of the combined vector can be solved to provide N mixtures Degree of methylation (eg methylation vector b). The M percent contribution may correspond to the vector x determined by solving for Ax=b. When N is greater than M, the solution may involve error minimization, such as using least squares. At step 170, the combined vector is used to determine the amount of each of the M tissue types in the mixture. The M value of the combined vector can be directly taken as the percentage contribution of the M tissue types. In some embodiments, the M value can be converted to a percentage. Error terms can be used to convert M values to higher or lower values.D . application As described above, percent contribution can be used for other measurements of biological samples and other determinations, such as whether a particular chromosomal region has sequence imbalance, whether a particular tissue type is diseased, and determining whether two parental haplotypes are inherited from the fetus of a pregnant woman from whom the sample is obtained Which haplotype is in the type. Figure 2 shows a schematic diagram showing several potential applications of DNA methylation deconvolution (eg, using plasma) according to embodiments of the present invention. In FIG. 2 , biological sample 205 is subjected to whole-genome bisulfite sequencing at 210 . At 230, plasma DNA tissue mapping uses tissue-specific methylation profiles 220 to determine percent tissue contribution. Example tissue-specific methylation profiles are shown for liver, blood cells, adipose tissue, lung, small intestine and colon. The percent contribution can be determined as described above and elsewhere, eg, solving for Ax=b. Examples of applications include prenatal testing 241 , cancer detection and monitoring 242 , organ transplant monitoring and organ damage assessment 244 . A list of methylation markers (genomic loci) suitable for determining the contribution of different organs to plasma DNA can be obtained by comparing different tissues, including liver, lung, esophagus, heart, pancreas, sigmoid colon, small intestine, adipose tissue, adrenal gland, Methylation signatures of colon, T cells, B cells, neutrophils, brain and placenta were identified (Figure 2). In various examples, genome-wide bisulfite sequencing data for liver, lung, esophagus, heart, pancreas, colon, small intestine, adipose tissue, adrenal gland, brain, and T cells were obtained from Baylor College of Medicine ) of the Human Epigenome Atlas (www.genboree.org/epigenomeatlas/index.rhtml). Bisulfite sequencing data for B cells and neutrophils from the publication of Hodges et al. (Hodges et al; Directional DNA methylation changes and complex intermediate states accompany lineage specificity in the adult hematopoietic compartment. Mol Cell 2011; 44: 17-28). Bisulfite sequencing data for the placenta were obtained from Lun et al. (Lun et al. Clin Chem 2013; 59:1583-94). In other embodiments, markers can be identified from data sets generated using microarray analysis, eg, using the Illumina Infinium HumanMethylation450 BeadChip array. II. Selection of methylation markers Above, we have described the principles of using methylation analysis to determine the composition of DNA mixtures. Specifically, the percent contribution of different organs (or tissues) to plasma DNA can be determined using methylation analysis. In this section, we additionally describe methods of selecting methylation markers and the clinical application of this technique. The results of determination of the composition of the DNA mixture by methylation analysis are influenced by the methylation markers used for deconvolution of the composition of the DNA mixture. Therefore, selection of appropriate genomic methylation markers can be important for accurate determination of the composition of DNA mixtures.A . Criteria for methylated markers for deconvolution For marker selection, the following three attributes can be considered: (i) methylated markers need to have low variability in the degree of methylation measured in the same tissue type across different individuals. Since determination of the composition of DNA mixtures depends on the identification of tissue-specific methylation patterns, the low variability in methylation levels in the same tissue type across different individuals would be suitable for accurately identifying tissue-specific patterns in DNA mixtures. In embodiments where tissue-specific methylation levels are obtained from samples from other organisms (eg, from a database), low variability means that the methylation levels from other samples are tissue-specific for the organism currently being tested The degree of methylation is similar. (ii) Methylation markers need to have high variability in the degree of methylation across different tissues. For a particular marker, higher differences in methylation levels across different tissues may provide a more precise determination of the contribution of different tissues to the DNA mixture. In particular, improvements in accuracy can be obtained by using one family of markers with attribute (ii) and another set of markers with attribute (iii). (iii) Methylation markers need to have a degree of methylation that is specifically different in a particular tissue compared to most or all other tissues. Compared to point (ii) above, a marker may have low variability in the degree of methylation in most tissues, but its degree of methylation in one particular tissue is different from that in most other tissues. This marker would be particularly useful for determining the contribution of tissues with different degrees of methylation from other tissues.B . example The rationale for marker selection is illustrated in Table 1 in the following hypothetical example. marker 1 marker 2 marker 3 marker 4 marker 5 marker 6 liver 1 20% 69% 9% 9% 10% 90% liver 2 50% 70% 10% 10% 10% 90% liver 3 90% 71% 11% 11% 10% 90% heart 20% 20% 30% 13% 12% 12% lung 30% 30% 60% 17% 14% 84% colon 40% 40% 90% 20% 80% 80% surface 1 . Methylation density in different tissues for 6 putative methylation markers. In this hypothetical example, when compared to marker 1, marker 2 has lower variability in methylation density in livers from three individuals. Thus, marker 2 is superior to marker 1 as a signature for determining the contribution of liver in the DNA mixture. Compared to marker 4, marker 3 has higher variability in methylation density across different tissue types. According to the mathematical relationship described above, the same degree of variation in the estimated contributions from different tissues will provide a greater variation in the inferred methylation density of the DNA mixture for marker 3 compared to marker 4. Thus, the assessment of the contribution of each tissue can be more precise in the case of marker 3. Marker 5 has low variability in methylation density across liver, heart and lung. Its methylation density varies from 10% to 14%. However, the methylation density of the colon is 80%. This marker would be particularly useful for determining the contribution of colon in DNA mixtures. Similarly, the heart was hypomethylated compared to other tissues for marker 6. Therefore, the contribution of the heart can be precisely determined by the marker 6 . Thus, the combination of markers 5 and 6 will enable accurate determination of colonic and cardiac contributions. Adding markers 2 and 3 would then be sufficient to infer the contribution of each of the four organs including liver, heart, lung and colon.C . different types of markers Methylation markers may not necessarily have all three of the above properties. Type I methylation markers will generally have properties (iii) above. A variety of such markers may also have attribute (i). Type II methylation markers, on the other hand, will generally have the above property (ii). A variety of such markers may also have attribute (i). It is also possible that a particular marker can have all three properties. In some embodiments, markers are broadly classified into two types (Type I and Type II). Type I markers are tissue specific. The degree of methylation of these markers on one or more tissues of a particular group differs from that of most other tissues. For example, a particular tissue may have a significant degree of methylation compared to the degree of methylation of all other tissues. In another example, two tissues (eg, tissue A and tissue B) have similar methylation levels, but the methylation levels of tissues A and B are significantly different from the methylation levels of the remaining tissues. Type II markers have high inter-tissue methylation variability. The degree of methylation of these markers is highly variable across different tissues. A single marker in this class may not be sufficient to determine the contribution of a particular tissue to the DNA mixture. However, combinations of Type II markers, or in combination with one or more Type I markers, can be used together to infer the contribution of individual tissues. Under the above definition, a particular marker may be a Type I marker only, a Type II marker only, or both Type I and Type II markers.1. I type Mark In one embodiment, type I markers can be identified by comparing the mean and standard deviation (SD) of the methylation density of the marker to the methylation density of this particular marker for all candidate tissues. In one embodiment, a marker is identified if its methylation density in one tissue differs from the mean of all tissues by 3 standard deviations (SD). The methylation profiles of 14 tissues obtained from the sources mentioned above were studied for marker selection. In one analysis, a total of 1,013 Type I markers were identified using the above criteria (markers labeled Type I in Table S1 of Appendix A of US Provisional Application No. 62/158,466). In other embodiments, other thresholds between a particular tissue and average methylation density may be used, such as, but not limited to, 1.5 SD, 2 SD, 2.5 SD, 3.5 SD, and 4 SD. In another embodiment, Type I markers can be identified by comparing the methylation density of a particular tissue to the median methylation density of all tissues. In other embodiments, when more than one tissue (eg, but not limited to, two, three, four, or five tissues) exhibits a methylation density that is significantly different from the average methylation density of all candidate tissues, Type I markers are available. In one embodiment, the cut-off methylation density can be calculated from the mean and SD of the methylation density of all candidate tissues. For illustrative purposes, the threshold can be defined as 3 SD above or below the mean methylation density. When more than one (eg, but not limited to, two, three, four, five, or more than five) tissues have methylation densities greater than 3 SD above the average methylation density of the tissues or below the average methylation density Select markers when the densities exceed 3 SD. 2. Type II markers To identify type II markers, the mean and SD of methylation densities across all 14 candidate tissues were calculated and the ratio of SD to mean was expressed as coefficient of variation (CV). In this illustrative example, we used a threshold of >0.25 for CV to identify eligible Type II markers, and the difference between the maximum and minimum methylation densities of tissue populations exceeding 0.2. Using these criteria, 5820 Type II markers (markers labeled Type II in Table S1 of Appendix A) were identified. In other embodiments, other thresholds may be used for CV, such as (but not limited to) 0.15, 0.2, 0.3, and 0.4. In other embodiments, other thresholds for the difference between maximum and minimum methylation densities may be used, such as, but not limited to, 0.1, 0.15, 0.25, 0.3, 0.35, 0.4, 0.45, and 0.5. In other embodiments, the average of multiple samples across the same tissue type can be used to measure changes in methylation levels across different tissues. For example, 10 methylation levels of the same genomic locus from 10 samples can be averaged to obtain a single methylation level for a genomic locus. Similar methods can be performed to determine the average degree of methylation for other tissue types of genomic loci. Averages across tissue types can then be used to determine whether genomic loci have significant variation across tissue types. Other statistical values other than the mean can be used, such as the median or geometric mean. Such statistics can be used to identify Type I and/or Type II markers. Different samples of the same tissue type (eg, from different individuals) can be used to determine changes in methylation levels across different samples. Thus, if there are multiple samples of the same tissue type, embodiments may additionally measure changes in specific markers between such samples of the same tissue type. A marker with low variation across the sample will be a more reliable marker than a marker with high variation. Additional details of markers and deconvolution can be found in co-owned US Patent Publication 2016/0017419 entitled "Methylation Pattern Analysis Of Tissues In A DNA Mixture" and "Non-Invasive Determination Of Methylome Of Methylation" by Chiu et al. In PCT Publication WO2014/043763 of Fetus Or Tumor From Plasma.D . different types of markers "Classes" of genomic loci (methylation markers) correspond to specific changes in the degree of methylation of loci across different individuals of the same tissue type. Different categories may have different ranges of variation between specific tissue types across individuals. Methylation markers of the first class may have a 10% or less difference in the degree of methylation between individuals tested. The second class of methylation markers can have greater than 10% differences in methylation levels between tested individuals. The use of methylated markers with low inter-individual variation (a first class of markers) would potentially improve the accuracy of determining the contribution of specific organs in DNA mixtures.E . Identify potential methylation markers In some embodiments, potential methylation markers are identified as follows. Such potential methylation markers can then be subjected to the above criteria to identify Type I and Type II markers. In other embodiments, there is no need to identify Type I or Type II. And other embodiments may use other techniques to identify potential methylation markers. In some embodiments, CpG islands (CGIs) and CpG banks on autosomes are considered for potential methylation markers. CGI and CpG banks on sex chromosomes were not used to minimize changes in methylation levels associated with sex-related chromosomal dosage differences in the source data. CGI was downloaded from the University of California, Santa Cruz (UCSC) database (genome.ucsc.edu/, 27,048 CpG islands on the human genome) (Kent et al., UCSC Human Genome Browser, Genome Res. 2002;12 (6):996-1006) and CpG shores are defined as 2 kb side windows of CpG islands (Irizarry et al. The human colon cancer methylome shows similar hypo- and hypermethylation at conserved tissue- specific CpG island shores. Nat Genet 2009; 41 ( 2): 178 - 186). Next, CpG islands and shores were subdivided into non-overlapping 500 bp units and each unit was considered a potential methylation marker. The methylation densities (ie the percentages of methylated CpGs within 500 bp units) of all potential loci were compared between the 14 tissue types. As previously reported (Lun et al. Clin Chem. 2013; 59: 1583-94), the placenta was found to be globally hypomethylated when compared to the rest of the tissue. Therefore, the methylation profile of the placenta was not included in the marker identification stage. Using the methylation profiles of the remaining 13 tissue types, two types of methylation markers were identified. For example, a type I marker can refer to any genomic locus that has a methylation density below or above 3 SD in a tissue when compared to an average of 13 tissue types. When (A) the methylation density of the highest methylated tissue is at least 20% higher than that of the lowest methylated tissue; and (B) the methylation density across 13 tissue types is divided by the mean of the group Type II markers are considered highly variable when the SD of methylation density (ie, the coefficient of variation) is at least 0.25. Finally, to reduce the number of potentially redundant markers, only one marker can be selected in a contiguous stretch of two CpG banks flanking a CpG island.F . Application-Based Selection The set of methylation markers selected for a particular application can vary depending on the parameters of the desired application. For example, for applications focused on haplotype or allele analysis, a suitable marker would be one of the heterozygous alleles on the same episomal DNA molecule. Since free DNA molecules (eg, plasma DNA) are typically less than 200 bp, suitable markers may be CpG sites within 200 bp hybrid loci (eg, SNPs). As another example, for applications where the release of DNA from a particular tissue into plasma is of particular importance, we may choose to prefer a larger number of methylation markers, where the tissue is Differentially methylated among types (eg type I markers). The number and selection of methylation markers in the deconvolution analysis can vary depending on the intended use. If the percent contribution of the liver is of particular interest, such as in patients who have undergone liver transplantation, more type I liver-specific markers can be used in deconvolution analysis to increase the accuracy of the quantification of the contribution of transplanted liver to plasma DNA . III. Compositional Accuracy As described above, the Examples can identify tissue contributors to plasma DNA. In various examples, genome-wide bisulfite sequencing of plasma DNA was performed and analyzed with reference to methylation profiles of different tissues. Using quadratic programming as an example, plasma DNA sequencing data was deconvolved into proportional contributions from different tissues. For pregnant women; patients with hepatocellular carcinoma, lung cancer and colorectal cancer; and individual test examples after bone marrow and liver transplantation. In most individuals, leukocytes are the major contributors to the circulating DNA pool. The placental contribution in pregnant women correlates with the proportional contribution as shown by fetal-specific genetic markers. Transplant-derived contributions to plasma in transplant recipients correlate with those determined using donor-specific genetic markers. Patients with hepatocellular, lung or colorectal cancer showed higher plasma DNA contributions from organs with tumors. The liver contribution in hepatocellular carcinoma patients was also correlated with measurements using tumor-associated copy number variation. In cancer patients and pregnant women exhibiting copy number variation in plasma, methylation deconvolution identifies biased tissue types. In pregnant women diagnosed with follicular lymphoma during pregnancy, methylation deconvolution indicates a generally higher contribution from B cells to the plasma DNA pool and localized B cells (rather than the placenta) as observed in plasma source of copy number variation. Thus, the embodiments may serve as powerful tools for evaluating a wide range of physiological and pathological conditions based on the identification of the perturbed proportional contributions of different tissues to plasma.A . Contribution of different types of blood cells As an example of methylation deconvolution, we determined the contribution of different tissues and cell types to circulating DNA. Two blood samples were collected from two patients with systemic lupus erythematosus (SLE). After collection, venous blood samples were centrifuged at 1,500 g for 10 minutes. After centrifugation, blood cells and plasma are separated. DNA is then extracted from the blood cells. DNA was bisulfite converted and sequenced using one lane of the flow cell in the HiSeq2000 sequencer. Two blood cell samples were analyzed using cell type specific methylation pattern analysis. The methylation patterns of neutrophils, lymphocytes, esophagus, colon, pancreas, liver, lung, heart, adrenal gland and hippocampus are included as potential candidates for blood cell DNA. 609 Methylation markers were selected for analysis. Whole blood samples from both individuals were also sent for cytometry to determine the neutrophilic and lymphocyte fractional composition of blood cells. blood sample 1 blood sample 2 Cell-type-specific methylation pattern analysis blood count Cell-type-specific methylation pattern analysis blood count Neutrophils 90.5% 93.6% 89.4% 89.9% lymphocytes 9.5% 6.4% 10.6% 10.1% esophagus 0% - 0% - colon 0% - 2% - pancreas 0% - 0% - liver 0% - 1% - lung 1% - 1% - heart 0% - 3% - adrenal glands 0% - 0% - hippocampus 0% - 0% - surface 2 . Blood Tissue Contribution by Deconvolution Pattern Analysis and Cell Counting For methylation pattern analysis, neutrophils and lymphocytes were determined to be the major components that make up blood cell DNA. The relative proportions of neutrophil and lymphocyte contributions were similar to their relative abundances in blood samples analyzed by cytometry.B . pregnant woman Methylation analysis of plasma DNA from pregnant women was used to analyze the contribution of different tissues, including liver, lung, pancreas, colon, hippocampus, small intestine, blood cells, heart, adrenal gland, esophagus and placenta. Since the placental genotype is generally the same as the genotype of the fetus but different from that of the pregnant woman, the precise contribution of the placenta to maternal plasma can be accurately determined by counting the number of fetal-specific alleles in a sample control group.1. composition and fetus DNA Percent Correlation Whole genome bisulfite sequencing of plasma DNA was performed on 15 pregnant women (five from each of the first, second and third trimesters). Methylation deconvolution was performed and percentage contributions from different tissues were inferred. The contribution of different organs was determined based on the degree of methylation (eg, methylation density) of all type I and type II markers in Table S1 using quadratic programming analysis. 3A shows a graph 300 of the percent contribution of different organs to plasma DNA for 15 pregnant women, according to an embodiment of the present invention. Each bar corresponds to the results for one sample. Different colors indicate the contribution of different organs to plasma. These results show that white blood cells (ie, neutrophils and lymphocytes) are the most important contributors to the plasma DNA pool. This observation is consistent with previous observations obtained after bone marrow transplantation ((Lui YY et al. Clin Chem 2002; 48:421-7). Figure 4 shows a table 400 for determining the percent contribution from tissue mapping analysis of plasma DNA in pregnant women, in accordance with embodiments of the present invention. These results also show that the placenta is another key contributor to plasma DNA in pregnant women, with fractional concentrations ranging from 9.9% to 38.4%. We also measured placental contribution using paternally inherited fetal single nucleotide polymorphism (SNP) alleles not possessed by pregnant females as previously described (31). To analyze fetal-specific SNP alleles, the fetus is genotyped by analyzing a chorionic villus sample or placenta. The genotype of pregnant women is determined by analyzing blood cells. SNP-based results show independent validation of methylation deconvolution results. Figure 3B shows a graph 350 of the correlation between the plasma DNA fraction inferred from the placental contribution of plasma DNA methylation deconvolution and the fetal DNA percent concentration inferred using fetal-specific SNP alleles according to embodiments of the invention . Graph 350 shows that placental contribution determined by methylation deconvolution has a strong correlation with percent fetal DNA concentration measured using SNPs (r=0.99, p<0.001, Pearson correlation). Therefore, a good positive correlation was observed between the values of the two parameters, indicating that plasma DNA methylation deconvolution accurately determines the contribution of the placenta to the maternal plasma sample. 5 shows a graph of the percent contribution of organs other than the placenta mapped from plasma DNA tissue and the percent fetal DNA concentration based on fetal-specific SNP alleles according to embodiments of the present invention. The X-axis represents the percent fetal DNA concentration estimated by SNP-based analysis and the Y-axis represents the percent contribution inferred by plasma tissue DNA mapping analysis. The plasma DNA contribution of neutrophils showed an inverse correlation. This may be due to the fact that neutrophils are the major contributors to the plasma DNA pool and thus placental contribution increases, the relative contribution from neutrophils will necessarily decrease. The methylation deconvolution results for the remaining tissues did not show a correlation with the percent fetal DNA concentration. 6 shows a table 600 of percent contribution from tissue mapping analysis of plasma DNA for a group of non-pregnant healthy control individuals according to embodiments of the present invention. When the method was applied to plasma from non-pregnant healthy controls, placental contribution was absent in most samples (median: 0%; interquartile range: 0% to 0.3%).2. Comparison of selected and random markers The precision of the percent contribution was tested by the selected marker relative to the random marker. Different compositional calculations were performed for different marker groups. One group was selected based on the criteria mentioned above, and the other group was a random group. The results show that it is important to unbiasedly select methylation marker (locus) uses in order to obtain accurate results. Eleven pregnant women and four healthy non-pregnant individuals were recruited for this analysis. Its plasma DNA was bisulfite converted and sequenced using an Illumina HiSeq2000 sequencer. Each plasma sample was sequenced by one channel of the sequencing flow cell. The sequence reads were then analyzed using the bioinformatics program Methy-Pipe (Jiang P. PLoS One 2014; 9: el00360). This program can align bisulfite-converted sequence reads to a reference genome and determine the methylation status of each CpG site on each sequenced fragment. The first set of markers is highly specific for identifying different tissues in plasma DNA. For each tissue type, the marker with the greatest difference in methylation density compared to other tissues was selected. The markers are determined from regions of the genome that contain at least one CpG dinucleotide. In this example, CpG islands (CGIs), which have high frequency of CpG sites in specific stretches of DNA, were used as potential markers. The CGI in this particular example was downloaded from the University of California, Santa Cruz (UCSC) repository: (genome.ucsc.edu). In total, we obtained 27,048 CpG islands from the human genome. The median size of the CpG islands was 565 bp (range: 200 bp to 45 kb). 90% of islands are less than 1.5 kb. For each methylation marker, the difference in methylation density between the tissue of interest and other tissues is determined. Differences are then expressed as the number of standard deviations (SD) across other tissues. All markers were graded according to this difference in methylation density for the tissue of interest. The 20 markers with the largest difference above (10 markers) and below (10 markers) the mean methylation density of other tissues were selected. The number of markers can vary, such as, but not limited to, 5, 15, 20, 30, 40, 50, 100, and 200. In addition, markers with high variability across all different tissues were also selected. In this example, markers with >50% difference between tissues with the highest and lowest methylation densities were selected. In other applications, other values may be used, such as, but not limited to, 20%, 30%, 40%, 60%, 70%, and 80%. In addition, the variability in methylation density across different tissues was also calculated based on the mean and SD. In this example, markers were also selected if the value of SD was more than twice the mean. In other applications, other thresholds, such as, but not limited to, 1, 1.5, 2.5, and 3 may also be used. Based on these selection criteria, 344 methylation markers were selected for the first set. For the second set, 341 markers were randomly selected from the 27,048 CGIs described above. All CGIs are first numbered 1 to 27,048. Next, random numbers (between 1 and 27,048) were generated by the computer for marker selection. This method was then repeated until a total of 341 markers were selected. If the generated random number is used, another random number will be generated. This set of markers is expected to have much lower specificity in identifying tissue-specific methylation patterns. Therefore, the accuracy of determining the composition of plasma DNA is expected to decrease. 7 shows a table 700 for the estimated contribution of different organs to plasma DNA for 11 pregnant females and 4 non-pregnant healthy individuals using the first set of markers (with high organ specificity) according to embodiments of the present invention. The percent fetal DNA concentration was determined by counting fetal-specific alleles and is shown in the bottom column. In each of the four non-pregnant control individuals, the contribution of the placenta to plasma was determined to be close to 0%. This indicates the specificity of this method. Figure 8 shows a table 800 for the estimated contribution of different organs to plasma DNA for 11 pregnant females and 4 non-pregnant healthy individuals using a second set of markers (with low organ specificity) according to embodiments of the present invention. The percent fetal DNA concentration determined by counting fetal-specific alleles is shown in the bottom column. Using these lower specific markers, a relative non-uniform percent contribution from the placenta was observed, and a considerable contribution from the placenta was observed in four non-pregnant control individuals. The tissue specificity of this indicator marker is important in this method. 9A is a graph 900 showing the correlation between estimated fetal DNA percent concentration (contributed from the placenta) and fetal DNA percent concentration determined by counting fetal-specific alleles in a maternal plasma sample. Using the first set of methylation markers, the results from the two techniques correlate well. Using the second set of methylation markers, however, evaluation by using methylation analysis showed significant deviations from the true value determined using fetal-specific allele counts. Figure 9B is a graph 950 showing the absolute difference between the assessment from methylation markers and the percent fetal DNA concentration determined by fetal-specific allele counting. The median error of the estimates using methylation analysis was 4% and 8% using the first and second set of markers, respectively.C . The impact of different standards As described above, various criteria can be used to identify different types of markers. For example, a type I marker can be identified by the degree of methylation in a particular tissue that differs from the average degree of methylation in all tissues (eg, by at least a certain threshold, such as 3 SD). And for Type II markers, a criterion of a certain change and maximum difference was used. The following sections show the accuracy of different standard identification markers. 1. Marker efficacy under less stringent criteria We compared the performance of methylation deconvolution analysis using markers with different variability across different tissues. The contribution of the placenta to plasma DNA was determined for 15 pregnant women based on two sets of markers with different selection criteria. Both sets of markers include all type I markers as described in the previous section. However, the selection criteria for Type II markers differed between the two groups of markers. Group I markers included all 5,820 type II markers that met the criteria of having a methylation density CV of >0.25 and the difference between the maximum and minimum methylation densities of the tissue population exceeding 0.2. For Group II markers, the CV requirement is >0.15 and the difference between the maximum and minimum methylation densities of the tissue population exceeds 0.1. There are 8,511 type II markers in this panel of markers. Figure 10A is a graph 1000 showing the inferred contribution of the placenta to plasma DNA using markers with different selection criteria according to embodiments of the present invention. The vertical axis corresponds to the placental contribution inferred using Group II markers. The horizontal axis corresponds to the placental contribution inferred using Group I markers. There was a good correlation (r=0.99, Pearson's correlation) between placental contribution outcomes based on two sets of markers with different selection criteria. Therefore, good precision can be obtained using the requirements of CV > 0.15 and the difference between the maximum and minimum methylation densities of the tissue population exceeding 0.1. 2. The effect of changes in methylation levels within the same tissue type To investigate whether changes in methylation of markers between the same tissue type (eg, from different individuals) would affect the performance of the deconvolution analysis, we analyzed placental tissue from two pregnancy cases. Two classes of methylation markers were identified. Specifically, the two classes were identified based on the similarity of their methylation levels in the two placental tissues. Markers of class i have a methylation density of 10% or less. Markers of class ii had high variability between the two placental tissues (difference in methylation density greater than 10%). Figure 10B is a graph 1050 showing the accuracy of plasma DNA deconvolution using markers with low variability (class i) and high variability (class ii) in the same tissue type. Plasma DNA deconvolution was performed to determine placental contribution to plasma DNA for 15 pregnant women. For each marker, the average of the methylation densities of the two placental tissues was used to represent the degree of methylation of the placentas under analysis. For each of the deconvolution analyses using class i and class ii markers, a total of 1,024 markers were used. The amount of placenta-derived DNA in plasma was also determined based on the ratio of fetal-specific SNP alleles. Percentage contributions inferred by methylation deconvolution analysis based on class i and class ii markers were then compared to results based on fetal-specific SNP alleles. The median deviation of derived placental contributions from values estimated based on fetal-specific alleles was 2.7% (using class i markers) and 7.1% (using class ii markers), respectively. Therefore, using class i markers with lower inter-individual variation in tissue methylation levels gave better accuracy in methylation deconvolution analysis. Significantly higher differences between results from methylation deconvolution and fetal-specific allele analysis were observed when markers with high variability within the same tissue type (class ii) were used (P<0.0001, Wilcoxon sign-rank test). In other words, using markers with low variability within the same tissue type will increase the precision of methylation deconvolution analysis. Thus, markers can be selected based on variability within the same tissue type, such as, but not limited to, CV values and the difference between maximum and minimum methylation densities within the same tissue type.IV. Deconvolution of Fetal Signatures If genomic signatures (eg, specific SNP alleles) are known, embodiments can determine which tissues are the source of such signatures. Thus, if a particular signature represents a fetus (eg, a paternal allele at a particular locus), the percentage contribution of the signature to placental tissue will be substantial. To demonstrate that single nucleotide changes can also be used to determine the source tissue from which the changes were derived, we analyzed plasma DNA from pregnant women. The placenta and maternal leukocytes were genotyped to identify SNPs for which the mother was homozygous and the fetus was heterozygous. We designate the allele shared by the fetus and mother as A and the fetus-specific allele as B. Thus, the mother has the genotype of AA and the fetus has the genotype of AB at each of these SNPs. Following bisulfite sequencing of maternal plasma DNA, all DNA fragments carrying a fetal-specific allele (B allele) and at least one CpG site were selected and used for downstream analysis. A total of 1.31 billion fragments were sequenced and 677,140 fragments carrying a fetal-specific allele (B allele) were used for deconvolution analysis. All CpG sites covered by at least 10 DNA fragments were used for deconvolution analysis. Other numbers of DNA fragments covering the sites can be used, such as 5, 15, 20, 25 or 30. Since the B allele is fetal-specific, these DNA fragments are expected to be derived from the placenta. organization Contribution ( % ) liver 0.9 lung 0.0 colon 0.0 small intestine 0.0 pancreas 0.5 adrenal glands 0.0 esophagus 3.1 Adipose tissue 0.0 heart 0.0 brain 0.3 T cells 0.0 B cells 0.0 neutrophils 0.0 placenta 95.2 surface 3 . Methylation deconvolution analysis using fetal-specific alleles. In Table 3, automethylation deconvolution analysis showed that the placenta was inferred to be the major contributor to these DNA fragments carrying fetal-specific SNP alleles. These results demonstrate that methylation deconvolution analysis accurately identifies the tissue origin of these DNA fragments carrying fetal-specific alleles. This shows that a particular allele can be assigned to the fetus. Such techniques are described in more detail below for determining fetal genotype and haplotype using methylation deconvolution analysis.V . Determination of fetal genome ( Mutation analysis ) Analyzing the inheritance of maternal mutations using maternal plasma DNA is a challenging task for non-invasive prenatal testing. For example, if a pregnant female is heterozygous for the mutation, analysis of fetal mutation status using maternal plasma DNA analysis will be technically difficult because both mutant and normal alleles will be present in the fetus regardless of the mutation status of her fetus. in its plasma. Previously, a number of different approaches have been developed to address this problem (Lun et al. Proc Natl Acad Sci USA. 2008;105:19920-5; Lo et al. Sci Transl Med. 2010;2:61ra91; Lam et al. Clin Chem. 2012; 58:1467-75). The rationale for these previous methods involved comparisons between the relative amounts of mutant and normal alleles in maternal plasma. To enhance the statistical power of the comparison, some of these methods additionally involve a comparison of the relative amounts of SNP alleles linked to mutant alleles and those linked to normal alleles. Alternatively or additionally, some embodiments of the invention may infer the mutational status of the fetus by methylation deconvolution analysis.A . Contribution of alleles using methylation deconvolution In this example, the genotype of the fetus is determined. The genotypes of the father and mother at a particular locus are assumed to be NN and MN, respectively. M and N indicate mutant and normal alleles, respectively. In this context, the fetus can inherit either the M allele or the N allele from the mother. Therefore, there are two possible genotypes in the fetus, MN and NN. In maternal plasma, the DNA carrying the fetal genotype is actually derived from the placenta. Thus, these DNA fragments will exhibit a placental methylation profile. Figure 11A shows a first scenario in which a fetus inherits the M allele from the mother and has the genotype of MN at a specific locus according to an embodiment of the invention. In the top portion of Figure 11A (genotypes marked), the father is shown with the genotype NN, the mother is shown with the genotype MN, and the fetus is shown with the genotype MN. DNA fragments exhibiting placental methylation profiles are marked with P, which is shown on fetal genotype. For example, a placental methylation profile can correspond to certain degrees of methylation at genomic loci proximate to a particular locus. DNA fragments aligned to a particular locus can also include genomic loci close to the locus (eg, within 200 bp of the locus), and thus can be used to measure the degree of methylation for methylation deconvolution analysis. Considering the genotype of the parent, the M allele is specific to the mother and the N allele is shared between the father and mother. In the bottom portion of Figure 11A (labeled maternal plasma), examples of two alleles M and N are shown, where each example represents a different DNA molecule in plasma at the locus of interest. For illustrative purposes, only a few DNA molecules are shown. In this example, the fetal DNA percentage is assumed to be 25%, as indicated by the P-labeling of 25% of the DNA molecules. In maternal plasma samples, we selectively analyzed DNA fragments carrying the M allele and performed methylation deconvolution analysis. Since the fetus has the genotype of MN, the placenta contributes both the M and N alleles to the maternal plasma DNA. Thus, some of the DNA fragments carrying the M allele will also carry placenta-specific methylation profiles at genomic loci close to the locus. Methylation deconvolution analysis will indicate that some of the DNA fragments carrying the M allele will be derived from the placenta, and thus the fetal genotype includes the M allele. Figure 11B shows a second scenario where the fetus inherits the N allele from the mother and has the genotype of NN at a specific locus according to an embodiment of the invention. In this case, only the DNA fragment carrying the N allele will exhibit a placental methylation profile in maternal plasma. Therefore, selective analysis of DNA fragments carrying the M allele by methylation deconvolution would indicate that these DNA fragments do not have a significant contribution from the placenta. Therefore, it can be determined that the fetus does not have M, and thus has the genotype of NN. In some embodiments, the placental contribution of the M and N alleles can be compared. Here, we assume that fetal DNA constitutes approximately 10% of the total maternal plasma DNA. Selective deconvolution of the M and N alleles would be useful to indicate which alleles the fetus has inherited from the mother. The expected results are shown in Table 4 below: fetal genotype MN NN Placental contribution of plasma DNA carrying the M allele roughly 10% Not significant (close to zero) Placental contribution of plasma DNA carrying the N allele roughly 10% roughly 20% Ratio of placental contribution of M and N alleles (M:N) 1:1 0:2 surface 4 . Placental contribution of M and N alleles to NN paternal genotype. In Table 4, the percent placental contributions of the M and N alleles can be compared. About the same placental contribution of both alleles (eg, within thresholds of each other) indicates that the fetal genotype is MN. On the other hand, a significantly higher placental contribution of the N allele compared to the M allele would be indicative of the fetal genotype of NN. In another embodiment, paternal genotype need not be taken into account. In this case, possible genotypes of the fetus include MM, MN and NN. fetal genotype MN NN MM Placental contribution of plasma DNA carrying the M allele roughly 10% Not significant (close to zero) roughly 20% Placental contribution of plasma DNA carrying the N allele roughly 10% roughly 20% Not significant (close to zero) Ratio of placental contribution of M and N alleles (M:N) 1:1 0:2 2:0 surface 5 . Placental contribution of M and N alleles for unknown paternal genotype. In Table 5, the placental contribution of DNA fragments carrying M and N alleles for different fetal genotypes is shown. When the fetus has the genotype of MM, the placental contribution of the M allele will be significantly higher than that of the N allele. When the fetus has the genotype of NN, the placental contribution of the N allele will be significantly higher than that of the N allele. When the fetus has the genotype of NM, the placental contribution of the M allele will be approximately equal to the placental contribution of the N allele. Thus, when the paternal genotype is unknown, the percent contribution of both alleles can be determined. That is, the first percent contribution can be determined using the first set of free DNA molecules aligned to the locus and including N. The degree of methylation of the first set of free DNA molecules can be measured at K genomic loci close to the locus. And a second percent contribution can be determined using a second set of episomal DNA molecules aligned to the locus and including M. The degree of methylation of the second set of free DNA molecules can be measured at K genomic loci close to the locus. For the first case where the fetal genotype is MN, the percent contributions determined for either allele will be approximately the same, as can be tested by determining whether the percent contributions are within thresholds of each other. To demonstrate the feasibility of this approach, we analyzed plasma DNA from pregnant women. Plasma DNA was bisulfite converted and analyzed using massively parallel sequencing. In addition, placental and blood cells were analyzed to determine fetal and maternal genotypes. For illustrative purposes, the analysis is located atKLF2 SNPs within genes. For this SNP, the genotypes of the mother and fetus are CG and CC, respectively. With this genotype combination, the placenta will contribute the C allele to maternal plasma, but all G alleles in maternal plasma will be derived from maternal tissue. In the sequencing data, there were 24 fragments carrying the G allele and 55 fragments carrying the C allele. CpG sites within these DNA fragments are used for methylation deconvolution. In this analysis, one goal was to determine the placental contribution of both alleles. To illustrate the principle, only placenta and blood cells were considered as candidate tissues for methylation deconvolution analysis. In another embodiment, three or more types of tissue may be used as candidates. In another embodiment, tissues are expected to have a significant contribution, eg blood cells, liver, lung, intestine and placenta may be used as candidates. C allele G allele C/G ratio placenta 62.6% 1.8% 34 blood cell 37.4% 98.2% surface 6 . Placental contribution of C and G alleles for unknown paternal genotype. In Table 6, the contribution from the placenta is inferred to be 62.6% and 1.8% for the C and G alleles, respectively. The placental contribution ratio of C/G was 34. These results indicate that the genotype of the fetus will be CC. This is consistent with the genotyping results of placental tissue. This example is different from and potentially more useful than previous methods of non-invasive prenatal testing based on allele ratio analysis of DNA with specific methylation patterns (Tong et al. Clin Chem 2006; 52: 2194-202) . In this aforementioned method, tissue-specific DNA is first identified from a DNA mixture (eg, plasma DNA) based on methylation patterns. For example, certain genes are completely unmethylated in blood cells and methylated in the placenta. Identification is performed using enzymes that keep methylated placental DNA intact. Therefore, all methylated DNA molecules remaining in plasma will be derived from the placenta and not from blood cells. Next, the allelic ratio of SNPs located on placenta-derived DNA molecules can be determined by measuring the amount of different alleles at the loci using intact placental DNA. When the fetus is heterozygous for the SNP, the ratio of the two alleles in the placenta-specific DNA will be approximately one. However, if the fetus is affected by an aneuploid chromosome and has three copies of the chromosome carrying this particular SNP, the ratio of the two alleles will be 1:2 or 2:1. In this aforementioned method, tissue-specific DNA molecules need to first be identified based on the methylation status specific to the tissue of interest. Methylated DNA molecules are unique to the placenta because blood cells are completely unmethylated as far as the target region is concerned. However, in embodiments of the present invention, the uniqueness of a certain methylation state is not required. Candidate tissues only need to differ in their methylation profiles, so more loci can be used, enabling haplotype deconvolution. Therefore, the tissue contribution of different alleles can be determined based on their methylation profiles. Additionally, the aforementioned methods may be more susceptible to statistical variation, since the number of fetal reads for each allele is directly compared to each other. However, when placental contributions are compared to each other, fetal read numbers are not directly compared to each other. Instead, the placental contribution is determined from all reads (methylated or unmethylated), and thus the placental contribution can be the same, even when the number of fetal reads is different. Thus, coverage bias for one haplotype can result.B . Determination of genetic haplotypes using deconvolution It has been previously shown that from the analysis of plasma DNA (or other cell-free DNA) of pregnant women carrying a fetus, the relative haplotype dose analysis (RHDO) method can be used to infer maternal haplotypes inherited from the fetus (Lo et al. Sci Transl Med). 2010; 2: 61ra91 and US Patent 8,467,976). In this method, we use haplotype information for pregnant women. This latter information can be obtained using methods of family analysis or direct analysis of haplotypes (eg Fan et al. Nat Biotechnol 2011; 29: 51-57; Snyder et al. Nat Rev Genet 2015; 16: 344-358). SNPs that are heterozygous in the mother but homozygous in the father can be used for RHDO analysis. Such use of specific SNPs can limit the loci that can be used, and thus limit the amount of data and precision. Embodiments may not be so limited to such specific SNPs. Additionally, the examples may be used in combination with the above references to provide increased precision. Example Methylation deconvolution can be used to determine placental contribution for two haplotypes using free DNA molecules. Placental contributions can be compared to determine which haplotype is inherited by the fetus. Examples may start with an inferred maternal or paternal haplotype, and then measure the degree of methylation of the plasma DNA molecules containing the SNP allele in each of their inferred haplotypes. We can then perform methylation deconvolution. Fetal haplotypes can be identified as the haplotype with the highest placental contribution from methylation deconvolution analysis. In all the above examples, paternal or maternal haplotypes can also be analyzed by family (ie, by analyzing the DNA of other family members) or by direct methods (such as those described by Fan et al. Nat Biotechnol 2012) Measured, not inferred haplotypes. 1. maternal haplotype In this example, we show plasma DNA methylation deconvolution analysis that can be used to infer maternal haplotypes inherited from an unborn fetus. Sources of genomic DNA, such as leukocyte DNA, from pregnant women can be subjected to genotyping, eg, using microarrays. The maternal genotyping results are then entered into a haplotype inference program (eg, IMPUTE2, Howie et al. PLoS Genet. 2009;7:e 1000529) to infer possible first and second maternal haplotypes. Population-specific genotype and haplotype information can be taken into account to improve the precision of inferences. In other embodiments, parental haplotypes can be analyzed by a single molecule, such as, but not limited to, by Fan et al. (Nat Biotechnol. 2011; 29:51-7), Kaper et al. (Proc Natl Acad Sci USA. 2013;110:5552-7), Lan et al. (Nat Commun. 2016;7:11784) and Selvaraj et al. (Nat Biotech 2013;31:1111-1118). Maternal plasma DNA can then be subjected to whole genome bisulfite sequencing and alignment to a reference genome sequence. Methylation deconvolution can then be performed for each of the predicted haplotypes. Since fetal DNA in maternal plasma is primarily derived from the placenta, the maternal haplotype inherited from the fetus is the haplotype showing the highest placental contribution. Maternal haplotype information can be used to link together SNP alleles and CpG sites on the same homologous chromosome. Next, SNP alleles can be used to identify DNA fragments from the same chromosomal replica (haplotype). CpG sites (or other sites) on this particular chromosome copy (haplotype) can be used for methylation deconvolution. Since the number of CpG sites available for deconvolution will be proportional to the number of SNPs on homologous chromosomes and is much larger than the number of CpG sites linked to a single SNP in haplotype-based deconvolution analysis, this method will More precise than deconvolution analysis using CpG sites linked to a single SNP. The principle is illustrated in Figure 12A. Figure 12A shows the determination of maternal haplotypes inherited from the fetus using methylation deconvolution in accordance with embodiments of the present invention. In the top part of Figure 12A, the two haplotypes of the mother and the fetus are shown at three loci and the mother is heterozygous. The two maternal haplotypes are labeled Hap I and Hap II. In this example, the fetus inherits Hap I from the mother. For illustrative purposes, only SNP loci for which the mother is heterozygous are shown. For illustration purposes, in this example, the father is homozygous for each of these loci. However, the same principle extends to the situation where the father is heterozygous without any change. In the bottom portion of Figure 12A (labeled maternal plasma), examples of two alleles at each locus are shown, where each example represents a different DNA molecule in plasma at the locus of interest. For illustrative purposes, only a few DNA molecules are shown. In this example, the fetal DNA percentage is assumed to be 20%, as indicated by the P-labeling of 20% of the DNA molecules. In maternal plasma, DNA molecules that carry the fetal genotype are derived from the placenta and thus carry placenta-specific methylation patterns. Circles marked with "P" indicate CpG sites exhibiting placental methylation patterns near the heterozygous locus. Reads including heterozygous loci and adjacent loci can be used to measure methylation levels to detect placental methylation patterns. In this example, one goal is to determine whether the fetus has inherited Hap I or Hap II from the mother. To achieve this goal, plasma DNA fragments carrying alleles on Hap I and covering at least one CpG site were selected for methylation deconvolution. Since the fetus inherits Hap I from the mother, the placenta will contribute a significant proportion of plasma DNA molecules to this pool. On the other hand, when fragments carrying alleles on Hap II were analyzed by methylation deconvolution, a very low contribution from the placenta would be observed. To illustrate this, we analyzed the maternal plasma samples stated above with respect to Table 6. We focused on the 5 Mb region on chromosome 1. SNPs for which the mother is heterozygous and the fetus is homozygous were selected for analysis. For each of these SNP loci, alleles shared between mother and fetus form one haplotype (denoted Hap I) and alleles present only on the maternal genome form the other haplotype type (denoted as Hap II). Thus, in this example, there are two maternal haplotypes (Hap I and Hap II) and the fetus inherits Hap I from the mother. In maternal plasma, DNA fragments carrying alleles on Hap I and those carrying alleles on Hap II were analyzed using methylation deconvolution, respectively. All CpG sites on the same plasma DNA molecule for hybrid SNPs were used for deconvolution analysis. In this example, none of these CpG sites overlapped with a Type I or Type II marker. Hap I Hap II liver 0% 0% lung 0% 6.7% colon 3.4% 6.2% small intestine 0% 10.6% pancreas 4.1% 0% adrenal glands 0% 4.6% esophagus 0% 0% Adipose tissue 3.7% 3.6% heart 0% 0% brain 6.8% 10.6% T cells 6.8% twenty one% B cells 8.9% 11.7% neutrophils 12.7% 25% placenta 53.5% 0% surface 7 . Methylation deconvolution of Hap I and Hap II. Table 7 shows the deconvolution of plasma DNA fragments carrying alleles on two maternal haplotypes, Hap I and Hap II. The fetus inherits maternal Hap I. From this deconvolution analysis, the placenta was inferred to contribute 53.5% of the plasma DNA fragments carrying the allele on Hap I. On the other hand, there is no placental contribution to the plasma DNA fragment carrying the allele on Hap II. Thus, methylation deconvolution analysis has accurately predicted fetal inheritance of Hap I from the mother. Greater precision can be achieved using CpG sites that overlap with Type I and/or Type II markers. As another example, to demonstrate the practical utility of this method, another pregnant woman was recruited. Obtain maternal peripheral blood. Blood samples were separated into plasma and cellular components. Maternal leukocytes were analyzed using the Illumina HumanOmni2.5-8 BeadChip array. We used IMPUTE2 (Howie et al. PLoS Genet. 2009;7:e1000529) to deduce the phase of 851 heterozygous SNPs on a 5 Mb region on the telomeric end of chromosome 1p. Haplotype phasing was based on a reference haplotype of 1000 genomes (mathgen.stats.ox.ac.uk/impute/1000GP_Phase3.tgz). After the phased haplotypes were obtained, CpG sites ligated to both haplotypes were used for methylation deconvolution. All CpG sites on the same plasma DNA molecule for hybrid SNPs were used for deconvolution analysis. In this example, none of these CpG sites overlapped with a Type I or Type II marker. Of the 851 SNPs used for deconvolution, 820 (96.2%) were on intronic and intergenic regions. None of them overlapped with CpG islands or shores. Hap I Hap II liver 0 0 lung 0 5.4 colon 0 6.2 small intestine 0 0 pancreas 0 25 adrenal glands 0 0 esophagus 0 0 Adipose tissue 0 17.8 heart 0 0 brain 0 0 T cells 11 7.9 B cells 0 0 Neutrophils 20.2 28.4 placenta 68.9 9.3 surface 8 . Methylation deconvolution of Hap I and Hap II. Table 8 shows the deconvolution of plasma DNA fragments carrying alleles on two maternal haplotypes inferred from a set of reference haplotypes. The two haplotypes were named Hap I and Hap II. It was inferred that Hap I had a significantly higher amount of placental contribution than Hap II, ie 68.9% versus 9.3%. Therefore, it is inferred that maternal Hap I has been inherited by the fetus. Maternal inheritance relying on haplotype inference is consistent with results from maternal and fetal genotypes. The advantage of this approach is that it is not limited to SNPs where the father of the fetus is homozygous and the mother of the fetus is heterozygous. Indeed, in the above examples, we have performed the analysis without knowing or inferring the paternal genotype or haplotype. This is an advantage over the methods described above ((Lo et al. Sci Transl Med 2010; 2: 61ra91; US Pat. No. 8,467,976; Fan et al. Nature 2012; 487: 320-324; Kitzman et al. Sci Transl Med 2012; 4: 137ra76) . In some embodiments, the first percent contribution of the first haplotype can be compared to a reference value derived based on the percent fetal DNA concentration to determine whether the haplotype is inherited by the fetus. Thresholds can be calculated as, for example, but not limited to, 1, 1.2, 1.4, 1.6, 1.8, 2, 2.2, 2.4, 2.6, or 2.8 times the percent fetal DNA concentration. In this way, if the first percent contribution is sufficiently large, the second percent contribution of the second haplotype need not be determined. In some embodiments, the inherited haplotype may have a deconvoluted fraction concentration that is double the fetal fraction and the non-hereditary haplotype has an insignificant contribution. The contribution of non-inherited haplotypes may not have a zero contribution, as paternal haplotypes may give noise to this analysis since some paternal alleles may be identical to maternal alleles. If the noise level is high, the percent contribution of the second haplotype can be determined, and the haplotype with the higher deconvolution fraction can be inferred to be inherited by the fetus. Some embodiments may test two haplotypes using a reference value to confirm that only one is inherited. If both appear to be inherited, the two percent contributions are comparable to each other. Additionally, if both appear to be inherited, the paternal genome can be tested, as the fetus may inherit a paternal haplotype that matches the non-inherited maternal haplotype. In other embodiments, the second percent contribution may be used to determine a reference value, eg, the second percent contribution plus a threshold value. Thus, the reference value may be the sum of the second percent contribution and the threshold value.2. paternal haplotype In another embodiment, methylation deconvolution analysis can be applied to the analysis of paternal haplotype inheritance. Figure 12B shows an illustration of paternal haplotype methylation analysis according to embodiments of the present invention. Methylation deconvolution can be performed on maternal plasma DNA fragments that carry alleles on paternal Hap III and Hap IV. Since the fetus has inherited Hap III, the placental contribution of Hap III will be higher compared to Hap IV. Therefore, the paternal inheritance of the fetus can be inferred. This embodiment has advantages over the aforementioned methods based on analysis of paternally specific alleles. For example, for SNP position 1, the A allele is present in the father, but not in the mother. Thus, detection of the paternal-specific A allele in maternal plasma indicates that Hap III is inherited by the fetus. However, for the SNP at position 2, neither the C nor T alleles were fetal-specific. In this case, paternal-specific allelic analysis cannot be used. However, the presence of paternal-specific alleles is not required for methylation deconvolution analysis. Therefore, SNPs heterozygous in both the father and mother can be used for methylation deconvolution analysis of the two paternal haplotypes. Thus, similar methods as used for maternal haplotypes can be used to determine which paternal haplotypes are inherited. In Figure 12B, the placental contribution of Hap III will be higher than that of Hap IV. The paternal haplotype can be determined in the same or similar manner as the maternal haplotype can be determined. 3. Using the method of deconvolution 13 is a flowchart illustrating a method 1300 for determining a portion of a fetal genome from a maternal sample using methylation deconvolution in accordance with embodiments of the present invention. Biological samples include mixtures of free DNA molecules from multiple tissue types, including maternal and fetal tissue types. A fetus has a father and a mother who is a pregnant woman. The portion of the fetal genome may be the entire chromosomal replica or only a portion of the chromosomal replica. The measured portions of the fetal genome can be combined to provide information about different portions of the fetal genome, up to the entire fetal genome. At step 1310, a plurality of free DNA molecules from the biological sample are analyzed. Step 1310 may be performed using the techniques described in step 140 of method 100 of FIG. 1 . For example, at least 1,000 free DNA molecules can be analyzed to determine where the free DNA molecules are located, and the degree of methylation can be measured as described below. In addition, the free DNA molecules are analyzed to determine the individual alleles of the free DNA molecules. For example, alleles of a DNA molecule can be determined from sequence reads obtained from sequencing or from specific probes hybridized to the DNA molecule, both techniques providing sequence reads (eg, probes can be obtained in the presence of hybridization) considered sequence reads). At step 1320, the first haplotype and the second haplotype of the first chromosomal region of the first parent genome of the first parent of the fetus are determined. Those skilled in the art will know various techniques for determining the haplotype of the parent. Haplotypes can be determined from the same sample as used to determine the degree of methylation below or from a different sample. In some embodiments, haplotypes can be determined from a sample of cells, such as the leukocyte layer of a blood sample or tissue of another organ. An example of determining haplotypes is provided in US Patent No. 8,467,976, which is incorporated herein by reference in its entirety. The first parent can be either the mother or the father. Other examples of methods of detecting parental haplotypes include, but are not limited to, those described by Fan et al. (Nat Biotechnol 2011; 29: 51-57), Snyder et al. (Nat Rev Genet 2015; 16: 344-358). Methods, GemCode technology from 10X Genomics (www.10xgenomics.com/) and Targeted Locus Amplification (TLA) technology from Cergentis (www.cergentis.com/). At step 1330, one or more heterozygous loci are identified from the first and second haplotypes. Each heterozygous locus has a corresponding first allele in the first haplotype and a corresponding second allele in the second haplotype. The one or more heterozygous loci can be the first plurality of heterozygous loci, wherein the second plurality of heterozygous loci can correspond to different chromosomal regions. At step 1340, a first plurality of free DNA molecules are identified. Each of the plurality of free DNA molecules is located at any of the heterozygous loci from step 1330 and includes the corresponding first allele, such that the free DNA molecule can be identified as corresponding to the first haploid type. It is possible that the episomal DNA molecule is located at more than one heterozygous locus, but typically a read will include only one heterozygous locus. Each of the first set of free DNA molecules also includes at least one of the N genomic loci, wherein the genomic locus is used to measure the degree of methylation. N is an integer, such as greater than or equal to 2, 3, 4, 5, 10, 20, 50, 100, 200, 500, 1,000, 2,000, or 5,000. Thus, reads of free DNA molecules can indicate coverage of 1 locus, 2 loci, etc. At step 1350, the methylation levels of the N first mixtures at N genomic loci (eg, CpG loci) are measured using a first set of a plurality of free DNA molecular weights. A first mixture degree of methylation can be measured for each of the N genomic loci. Step 1350 may be performed in a similar manner to step 150 of method 100 of FIG. 1 . In some embodiments, the degree of methylation of DNA molecules can be measured using methylation sensing sequencing results, which can also be used to determine the location and individual alleles of DNA molecules. Those skilled in the art will understand various techniques that can be used to determine the methylation status of sites on DNA molecules. At step 1360, the N first degrees of methylation are used to determine a first percent contribution of fetal tissue types in the mixture. In some embodiments, step 1360 may be performed via steps 160 and 170 of method 100 of FIG. 1 . Thus, the percent contribution can be determined simultaneously for a set of M tissue types. Step 1360 may use the N tissue-specific methylation levels for N genomic loci determined for each of the M tissue types, such as step 120 of method 100 of FIG. 1 . At step 1370, a second plurality of free DNA molecules are identified. Each of the plurality of episomal DNA molecules is located at any of the heterozygous loci from step 1330 and includes the corresponding second allele, such that the episomal DNA molecule can be identified as corresponding to the second haploid type. Each of the second set of free DNA molecules also includes at least one of the N genomic loci, wherein the genomic locus is used to measure the degree of methylation. At step 1380, the N second mixture of N genomic loci is measured for methylation degree using a second plurality of free DNA molecular weights. Step 1380 may be performed in a manner similar to step 1350. At step 1385, the N second degrees of methylation are used to determine the second percent contribution of fetal tissue types in the mixture. Step 1385 may proceed in a manner similar to step 1360. At step 1390, a first separation value between the first percent contribution and the second percent contribution is calculated. Examples of separate values are described herein, including, for example, differences or ratios. At step 1395, a portion of the fetal genome is determined at one or more heterozygous loci based on the first segregation value. Thus, the genetic haplotype of the first parent can be determined. For example, the first separation value may be the ratio of the first percent contribution to the second percent contribution. When the ratio is greater than a threshold value, a portion of the fetal genome can be determined to have one or more copies of the first haplotype and no copies of the second haplotype. Examples of threshold values include, but are not limited to, 1.3, 1.4, 1.5, 1.6, 1.8, 2.0, 2.2, 2.4, 2.6, 2.8, and 3.0. When the ratio is less than a threshold value, a portion of the fetal genome can be determined to have one or more copies of the second haplotype and no copies of the first haplotype. Examples of threshold values include, but are not limited to, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 0.8. When the ratio is equal to one within the threshold, the portion of the fetal genome can be determined to have the first haplotype and the second haplotype. Examples of thresholds include, but are not limited to, 0.85, 0.9, 0.95, 1.0, 1.05, 1.1, and 1.15. Two haplotypes can be inherited when two parents have the same haplotype in the region analyzed. As another example, the first separation value is the difference between the first percent contribution and the second percent contribution. When the difference is greater than a threshold value, a portion of the fetal genome can be determined to have one or more copies of the first haplotype and no copies of the second haplotype. Examples of threshold values include, but are not limited to, 1%, 1.5%, 2%, 2.5%, 3%, 4%, 5%, 6%, 7%, 8%, 10%, 12%, 14%, 16%, 18% and 20%. When the difference is less than a threshold value, eg, when the threshold value is negative, a portion of the fetal genome can be determined to have one or more copies of the second haplotype and no copies of the first haplotype. The genetic haplotype of the other parent can also be determined. For example, a second plurality of heterozygous loci for a first chromosomal region can be identified in the genome of another parent. The percent contribution can be determined for each of the haplotypes of the other parent, and the segregation value can be used to determine the genetic haplotype of the other parent. For example, the first plurality of heterozygous loci and the second plurality of heterozygous loci can be the same locus or different. Each of the second plurality of heterozygous loci can include the corresponding third allele of the first haplotype (eg, first paternal haplotype) set of the other parent and the second haplotype of the other parent The corresponding fourth allele in the plotype (eg, the second paternal haplotype). The third and fourth alleles can be the same as the first and second alleles. In addition to the first and second sets of free DNA molecules of the first parent, each of the third plurality of free DNA molecules may be located at any one of the second plurality of hybrid loci, including the corresponds to the third allele and includes at least one of the K genomic loci. The K genomic loci can be the same or different from the N genomic loci used for the first parent. In a similar manner to the first parent, the K third mixture methylation degree can be measured at the K genomic loci using the second plurality of free DNA molecules of the third set, and the K third methylation degree can be used The third percent contribution of fetal tissue types in the mixture was determined. The third percent contribution corresponds to the first haplotype of the other parent (eg, the first paternal haplotype). The fourth plurality of episomal DNA molecules can each be located at any one of the second plurality of heterozygous loci, including the corresponding fourth allele of the heterozygous locus, and including at least one of the K genomic loci one. Therefore, the fourth set of DNA can be used to test the second haplotype of another parent. The second plurality of free DNA molecular weights of the fourth group can be used to determine the degree of methylation of the K4 mixture of K genomic loci, and the degree of methylation of the fourth of the fetal tissue in the mixture can be determined using the K4 methylation degree. Four percent contribution. A second separation value between the third percent contribution and the fourth percent contribution can be calculated, and the portion of the fetal genome at the second plurality of heterozygous loci can be determined based on the second separation value. The genetic haplotype from the other parent can be determined in a similar manner as for the first parent. The fourth percent contribution corresponds to the second haplotype of the other parent (eg, the second paternal haplotype). In some embodiments, the second percent contribution need not be determined. Instead, if the corresponding percentage contribution is high enough, the haplotype can be determined to be inherited. For example, the first percent contribution can be compared to a reference value to determine whether the fetus has inherited the first haplotype at the first chromosomal region. When the first percent contribution exceeds the reference value, the fetus can be determined to inherit the first haplotype at the first chromosomal region. In other embodiments, the reference value can be determined from the second percent contribution. For example, the reference value may be the sum of the second percent contribution and the threshold value. The sum with the threshold value ensures that the first percent contribution is sufficiently larger than the second percent contribution. A separate determination of inheritance of the second haplotype can be performed by comparing the second percent contribution to a reference value to determine whether the fetus has inherited the second haplotype at the first chromosomal region. When the second percent contribution exceeds the reference value, the fetus can be determined to inherit the second haplotype at the first chromosomal region. If both percent contributions are determined to exceed a reference value, the two percent contributions can be compared to each other to determine whether one is significantly greater than the other (eg, using a threshold value). The haplotype of the other parent can be determined to identify whether one of these haplotypes is identical to the haplotype of the first parent, thereby explaining that both haplotypes of the first parent can be inherited.C . Determination of genetic haplotypes using methylation levels Other embodiments may use the general hypomethylation of cell-free fetal DNA to identify genetic haplotypes as those with one of the lower overall methylation levels. Examples may start with an inferred maternal or paternal haplotype, and then measure the degree of methylation of the plasma DNA molecules containing the SNP allele in each of their inferred haplotypes. In one embodiment of analyzing maternal haplotypes, the degree of methylation of two putative maternal haplotypes can be compared, and the one with the lower degree of methylation will be predicted to be the haplotype inherited by the fetus. In another embodiment of analyzing paternal haplotypes, the methylation levels of two putative paternal haplotypes can be compared, and the one with the lower methylation level will be predicted to be the haplotype inherited from the fetus type.1. Example For example, the degree of methylation of each of the two maternal haplotypes can be determined. Since placental tissue is relatively hypomethylated compared to other tissues, we expected that maternal haplotypes inherited from the fetus would be less methylated than haplotypes not inherited from the fetus. Methylation densities were tested in maternal plasma using the mother's actual haplotypes inferred using maternal, paternal and fetal genotypes. Hap I Hap II total methylation density 65% 87% surface 9 . Methylation densities of actual Hap I and Hap II. Table 9 shows the methylation densities of the two maternal haplotypes in maternal plasma. Since Hap I is the actual haplotype inherited from the fetus according to genotyping, the results of methylation analysis of the haplotypes properly identify inheritance. In other embodiments, maternal haplotypes can be inferred based on individual maternal genotypes, or reference haplotypes from a population of haplotype databases can also be used for this analysis. The maternal haplotypes used in this example were phased using the IMPUTE2 program. Therefore, inferred maternal haplotypes can also be used for this analysis. Hap I Hap II total methylation density 68% 76% surface 10 . Inferred methylation densities for Hap I and Hap II. Table 10 shows the methylation densities of the two putative maternal haplotypes in maternal plasma. Maternal haplotypes inferred from fetal inheritance have lower methylation densities. Examples of statistical procedures that can be used to determine whether a haplotype has a sufficiently low methylation density include the chi-square test. It may be desirable that the separation between the two methylation levels be sufficiently large (eg, greater than a threshold value) for the determination. If the separation is insufficient, indeterminate classification can be performed. In some embodiments, determination of the inheritance of the two haplotypes can determine whether the separation is not sufficiently large and whether both methylation levels are below a threshold level, which can be characterized by including fetal DNA. For example, Tables 9 and 10 indicate that methylation densities below 70% may indicate that the fetus has inherited the haplotype. When the parents share haplotypes for the region analyzed, two haplotypes can be inherited. In another example, the total methylation densities of maternal plasma DNA carrying paternal Hap III and Hap IV can be compared. Similar to maternal haplotype analysis, the fetus will be inferred to inherit the paternal haplotype with lower overall methylation density.2. Methods using methylation levels 14 is a flowchart illustrating a method 1400 for determining a portion of a fetal genome from a maternal sample using methylation levels according to embodiments of the present invention. Biological samples include mixtures of free DNA molecules from multiple tissue types, including maternal and fetal tissue types. A fetus has a father and a mother who is a pregnant woman. The portion of the fetal genome may be the entire chromosomal replica or only a portion of the chromosomal replica. The assayed portions of the fetal genome can be combined to provide the entire fetal genome, as with other methods described herein. At step 1410, a plurality of free DNA molecules from the biological sample are analyzed. Step 1410 may be performed in a similar manner to step 1310 of method 1300 of FIG. 13 . At step 1420, the first haplotype and the second haplotype of the first chromosomal region of the first parent genome of the first parent of the fetus are determined. Step 1420 may be performed in a similar manner to step 1320 of FIG. 13 . In some embodiments, the genome of the first parent can be genotyped at multiple heterozygous loci using a sample from the first parent, such as a blood sample or other tissue that may or may not contain fetal DNA. A plurality of reference haplotypes can be obtained, eg, from a database of reference genomes. The first and second haplotypes can be inferred using the genotype and the plurality of reference haplotypes. For example, the alleles of each genotype can be compared relative to a reference haplotype, and any haplotypes that do not include an allele at the corresponding locus can be discarded. Once the two reference haplotypes are retained, those haplotypes can be identified as the first and second haplotypes. At step 1430, a plurality of heterozygous loci are identified from the first and second haplotypes. Each heterozygous locus has the first allele in the first haplotype and the second allele in the second haplotype. At step 1440, a first plurality of free DNA molecules are identified. Step 1440 may be performed in a similar manner to step 1340 of FIG. 13 . At step 1450, the degree of methylation of the first mixture is measured using a first set of free DNA molecular weights. For example, the degree of methylation of the first mixture can be the methylation density of the first set of free DNA molecules. The methylation density can be calculated as the total methylation density of all free DNA molecules of the first group. In another example, separate methylation densities can be calculated for each locus, and separate methylation densities can be combined to obtain a first mixture degree of methylation, eg, an average of separate methylation densities. At step 1460, a second plurality of free DNA molecules are identified. 1460 may be performed in a similar manner to step 1370 of FIG. 13 . At step 1470, the degree of methylation of the second mixture is measured using a second set of free DNA molecular weights. For example, the degree of methylation of the second mixture can be the methylation density of the second set of free DNA molecules. At step 1480, it is determined which of the first haplotype and the second haplotype is inherited by the fetus based on which of the first and second mixture levels of methylation is lower. As part of step 1480, a separation value between the degree of methylation of the first mixture and the degree of methylation of the second mixture can be determined and compared to a threshold value. Threshold values ensure that the lower level is sufficiently low. Threshold values can be determined using the chi-square test. For example, measurements can be performed on samples of known genetic haplotypes, and the distribution of segregation values can be determined, and optionally, thresholds for accurate determination of genetic haplotypes in the training data obtained from the samples. Methods 1300 and 1400 can also be combined, where each method is performed in a check format, and the genetic haplotype determines whether the two methods are consistent with each other.D . locus selection Various embodiments can be used to compare the degree of methylation or percent contribution of two putative maternal haplotypes in maternal plasma. In one embodiment, the number of SNP loci to be analyzed can be determined prior to analysis. For example, determinations for haplotypes can be based on a variety of factors, such as, but not limited to, the statistical power required, the mean difference in methylation levels of the placenta and blood cells in the region of interest, and the number of molecules analyzed for each SNP. Number of SNP loci for deconvolution analysis. The size of the region of interest can be fixed and all SNPs within the region of interest can be used for analysis. Various factors can be considered, such as, but not limited to, the required statistical power, the mean difference in methylation levels of placenta and blood cells in the region of interest, the number of molecules analyzed for each SNP, and the probability of meiotic recombination with the region of interest , to measure the size of the region of interest. In other embodiments, the number of SNPs and the size of the region to be analyzed are not determined prior to analysis. For example, the number of SNPs can be sequentially increased until the data is sufficient to draw a statistically significant conclusion about which maternal haplotype is statistically significantly less methylated than the other. For example, SNPs on a region of interest can be arranged in ascending order of their genomic coordinates. Next, statistical tests can be performed with the data for the lowest number of SNPs with genomic coordinates. If this is sufficient to draw a conclusion about which haplotype is statistically less methylated, a conclusion is made. Similarly, SNPs can be sorted in descending order with the highest number of genomic coordinates sufficient to use. If statistical precision is insufficient, another statistical comparison can be made starting with the next SNP with a higher number of genomic coordinates. On the other hand, if the data for the first SNP is insufficient to conclude that one haplotype is less methylated than the other (or the separation value between the percentage contributions is not sufficiently large), the data for another SNP can be added And perform another round of statistical testing. This procedure can be continued until the accumulated data are sufficient to draw a statistically significant conclusion. Various statistical tests can be performed to compare the degree of methylation of two haplotypes, such as (but not limited to) Student's t-test, Mann-Whitney rank-sum test (Mann-Whitney rank-test) sum test) and chi-square test. The degree of statistical significance can be determined based on the desired confidence in the conclusion, for example, but not limited to, using a P-value of 0.05, 0.01, 0.001, 0.0001, or 0.00001.E . and RHDO Of combination In some embodiments, the results generated by the RHDO analysis of US Pat. No. 8,467,976 can be combined with the methylation embodiments of the present invention to achieve a more precise diagnostic procedure or to reduce the amount of sequencing required. For example, fetal haplotypes can be determined using the present examples and the results of the RHDO analysis using US Pat. No. 8,467,976, and fetal haplotypes determined from the two techniques can be compared. For example, the results of two analyses will only be acceptable if they agree. If the two analyses show different conclusions, other analyses can be performed, eg, the genome can be replicated at a higher depth of coverage. To make such combined methods most cost-effective, it is preferable to have a type of sequencing that can produce data for both methods. In one embodiment, this can be done by single-molecule methods that will generate sequencing and methylation information, such as using single-molecule real-time sequencing technology from Pacific Biosciences, or nanopore sequencing (eg, from Oxford Nanopore Technologies ). These are two examples of methylation sensing sequences. In another embodiment, RHDO analysis can be performed on bisulfite sequencing results. For such embodiments, any maternal and paternal genetic information can also be determined using bisulfite sequencing. Bisulfite sequencing is thus another example of methylation sensing sequencing. In addition, other methylation-sensitivity sequencing techniques can be used, such as oxidative bisulfite sequencing (Booth et al. Science 2012; 336:934-937) or Tet-assisted bisulfite sequencing (Yu et al. Cell 2012; 149: 1368-1380). The latter example will allow us to analyze the 5-methylcytosine distribution of the DNA molecules analyzed.F . Using knowledge of the fetal genome Non-invasive prenatal analysis of the fetal genome can be used to determine whether the fetus has inherited the disease from the parent. This is particularly useful for detecting monogenic disorders such as congenital adrenal hyperplasia (New et al J Clin Endocrinol Metab 2014;99:E1022-30), beta-thalassemia (Lam et al Clin Chem. 2012;58:1467) -75) and hereditary muscular dystrophy (Genet Med 2015;17:889-96). If a monogenic disorder is detected, various treatments can be performed, eg, pregnancy can be terminated, treatment provided before pregnancy or after birth. For example, steroid treatment can be administered prenatally to pregnant women confirmed to have fetuses affected by congenital adrenal hyperplasia to avoid abnormal sexual development.VI . Haplotype Deconvolution Analysis for Aneuploidy Detection Haplotype deconvolution can also be used to detect sequence imbalances in chromosomal regions of the fetus, such as aneuploidy, microdeletions, or microamplifications (eg, microduplications). For example, the percent contribution of a haplotype in one region can be compared to the percent contribution of another haplotype in another region.A . Mother 15 shows chromosomal aneuploidy detection based on haplotype deconvolution for maternal haplotypes according to embodiments of the present invention. In this illustration, the mother has two maternal haplotypes, Hap I and Hap II. For illustrative purposes, we assume that 80% of their plasma DNA is derived from its own cells and 20% is derived from the placenta, which are exemplary percentages within the ranges commonly measured. This method can be generally applied to pregnant women with different percentages of fetal DNA. Knowledge of fetal DNA percentage is not required, but is provided for illustration only, although measurement of fetal DNA percentage can be performed in a variety of ways, such as using fetal-specific alleles or fetal-specific methylation markers. The fetus inherits Hap I and the other haplotype is from the father, namely Hap III. Placenta-derived DNA will reveal fetal genotype and sequence imbalances can be detected by analyzing the percent contribution of placenta-derived DNA generated. As explained above, fetal inheritance of maternal haplotypes can be determined via deconvolution of the two maternal haplotypes. Analysis of the contribution of the placenta to maternal DNA can be performed for each of the two maternal haplotypes. The maternal haplotype inherited from the fetus (Hap I in this example) will have a much higher placental contribution than the maternal haplotype not inherited from the fetus (Hap II). The placental contribution of Hap I will be positively correlated with the percent fetal DNA concentration in maternal plasma. After determining which maternal haplotype is inherited by the fetus, the dose of chromosomes inherited by the fetus from the mother can be further determined via maternal haplotype deconvolution. In this illustration, two chromosomal regions are analyzed using maternal haplotype deconvolution. In one embodiment, the reference chromosome (RefChr) is a chromosome or chromosomal region that is unlikely to be affected by a chromosomal aneuploidy. Reference chromosomal regions are shown on the left side of FIG. 15 . A target chromosome (TargetChr) is a chromosome or chromosomal region potentially affected by a chromosomal aneuploidy. Target chromosomal regions are shown on the right side of Figure 15. The two regions can be different regions of the same chromosome or regions of two different chromosomes. In the example shown, it has been inferred from the methylation deconvolution of Hap I and Hap II for each region that the fetus has inherited Hap I from the mother for both the reference and target chromosomes. Next, the placental contribution of maternal plasma DNA for Hap I can be compared between the reference chromosome and the chromosome of interest. A sequence imbalance can be identified if the placental contribution of Hap I of the target chromosomal region is significantly different from the placental contribution of Hap I of the reference chromosomal region (eg, due to higher amplification or lower due to deletion). For illustrative purposes, we use the detection of trisomy sex as an example. However, other types of chromosomal aneuploidy (including single chromosomes), amplification of subchromosomal regions, or deletions of subchromosomal regions can also be detected using this method. For trisomies, additional copies of the affected chromosomes may be inherited from the father (denoted trisomies (F)) or mothers (denoted trisomies (M)). In more than 90% of cases of trisomy 21, the additional copy of chromosome 21 is derived from the mother (Driscoll et al. N Engl J Med 2009;360: 2556-2562). In the case of trisomy (M), the placental contribution of Hap I of the target chromosome will be higher than that of the reference chromosome. In Figure 15, the trisomy (M) is shown by two instances of Hap I, which will provide a higher placental contribution to the target region than one instance of Hap I to the reference region. Whether the placental contribution of Hap I to the target chromosome is higher than the contribution to the reference chromosome can be determined by comparing the separation value between the two placental contributions and a threshold value, which can be measured based on the percentage of fetal DNA. A higher percentage of fetal DNA will result in a higher expected separation value between the two placental contributions and thus the threshold value can be set higher. For example, with a fetal DNA percentage of 20%, the placental contribution of Hap I to the reference region will be about 20% and the placental contribution of Hap I to the target region will be about 36.4%. For example, given that 10 DNA molecules are present at the reference chromosome, two of them are fetal and eight of them are maternal. For the two fetal DNA molecules, one was derived from Hap I and one was derived from Hap III. Of the eight parent DNA molecules, four were Hap I and four were Hap II. For the target region, there will be additional DNA molecules of Hap I from the fetus. Thus, there would be a total of two molecules of fetal Hap I DNA and four molecules of maternal Hap I DNA, providing 2/6=33.3%. The threshold for the difference (eg, 13.3%) can be placed between 0 and 13.3% to provide optimal specificity and sensitivity. The distribution of isolated values can be determined from a reference sample set. In the context of euploidy, the placental contribution will be approximately the same, eg the segregation value will be less than the threshold value. Based on the descriptions herein and in US Pat. No. 8,467,976 and other references cited herein, one skilled in the art will know how to select an appropriate threshold value. In one embodiment, the ratio (or other segregation value) of the placental contribution of Hap I between the target and reference chromosomes of a population of pregnant women known to carry euploid fetuses can be used as the reference interval. The ratios in the test cases can be compared to this reference group to determine if there is a significant increase in the placental contribution of Hap I with respect to the target region relative to the reference region. In the example of 20% fetal DNA, the ratio would be 33.3/20=1.67. The ratio may generally be 2/(1+f), where f represents the percent fetal DNA concentration. In another embodiment, the difference in the placental contribution of Hap I between the target and reference chromosomes can be determined. This difference is then compared to the reference group.B . Father In another embodiment, haplotype deconvolution of the paternal haplotypes (Hap III and Hap IV) can be performed in maternal plasma. Analysis of paternal haplotypes can be performed in a similar manner as for maternal haplotypes. 16 shows chromosomal aneuploidy detection based on haplotype deconvolution for paternal haplotypes according to embodiments of the present invention. In this illustration, the father has two paternal haplotypes, Hap III and Hap IV. As in Figure 15, the fetus inherits Hap I from the mother and Hap III from the father. In the case where the extra copy of the chromosome is derived from the father (trisomy (F)), the placental contribution of Hap III to the target chromosome will be higher than to the reference chromosome. This is shown for the trisomy (F) example, where two copies of Hap III are shown. As described above with respect to maternal haplotypes, the separation value between the placental contributions of Hap III to the target and reference regions can be compared to a threshold value to determine whether additional copies of Hap III are present in the target region. In various embodiments, the ratio or difference of the two placental contributions of the test case can be compared to a reference group of each pregnant female known to carry an euploid fetus to determine whether the fetus has a chromosomal trisomy of the chromosome of interest, or Amplification or deletion of the target chromosomal region. Threshold values can be based on a reference group of euploid fetuses, a reference group of aneuploid fetuses, or a segregated value of the two. Separate measurements of fetal DNA percentage can also be used, as described herein.C . Methods of Detecting Sequence Imbalance 17 is a flowchart of a method 1700 for detecting sequence imbalances in a portion of the fetal genome of a pregnant woman's unborn fetus using a biological sample from a pregnant woman, according to an embodiment of the invention. At step 1710, a plurality of free DNA molecules from the biological sample are analyzed. Step 1710 may be performed in a similar manner to step 1310 of method 1300 of FIG. 13 . At step 1720, the first target haplotype is determined for the target chromosomal region of the first parental genome of the first parent of the fetus, and the first reference haplotype is determined for the reference chromosomal region of the first parental genome. Step 1720 may be performed in a similar manner to step 1320 of FIG. 13 . The target chromosomal region and the reference chromosomal region may be the entire chromosome or only a portion of the chromosome. Thus, the target chromosomal region can be a first chromosome and the reference chromosomal region can be a second chromosome different from the first chromosome. The first parent may be the mother or father of the fetus. The target chromosomal region can be selected based on various criteria. For example, a plurality of target regions may be selected, as it may be conceivable to test multiple non-overlapping regions of a given size, such as 1 Mb, 5 Mb, 10 Mb, 20 Mb, 50 Mb, etc. As another example, a target chromosomal region can be selected based on a copy number analysis that identifies the region as having more DNA molecules than expected, eg, as described in US Patent Publications 2009/0029377 and 2011/0276277. In some embodiments, it can be determined that the fetus has inherited the first target haplotype from the first parent and the fetus has inherited the first reference haplotype from the first parent. The assay may include the embodiment of FIG. 13 or FIG. 14 . For example, determining that the fetus has inherited the first target haplotype from the first parent may include determining the second target percent contribution of fetal tissue types in the mixture corresponding to the second target haplotype, calculating the first target percent contribution and the second target percent contribution. A second segregation value between the two target percent contributions and determining that the fetus inherits the first target haplotype from the first parent based on the second segregation value. At step 1730, a plurality of target heterozygous loci for the target chromosomal region of the first parental genome are identified. Each target heterozygous locus includes a corresponding first target allele in a first target haplotype species in a first chromosomal region of the first parental genome and a corresponding second target allele in a second target haplotype. Referring back to the example of Figure 15, the heterozygous locus of interest has a corresponding first target allele of {G,T,A} on Hap I and a corresponding second target of {A,G,C} on Hap II allele. At step 1740, a plurality of free DNA molecules of the target set are identified. Each episomal DNA molecule of the target set is located at any of the target heterozygous loci, including the corresponding first target allele, and including at least one of the N genomic loci in the target chromosomal region. Step 1740 may be performed in a manner similar to that described herein. For example, sequence reads can be mapped to a reference genome in which a plurality of episomal DNA molecules of a target set are aligned with any of the target hybrid loci. At step 1750, the methylation levels of the N first mixtures of N genomic loci are measured using the plurality of free DNA molecular weights of the target set. Step 1750 may be performed in a similar manner to step 1350 of FIG. 13 . At step 1760, the N first degrees of methylation are used to determine the first percent contribution of fetal tissue types in the mixture. Step 1760 may be performed in a similar manner to step 1360 of FIG. 13 . At step 1770, a plurality of reference heterozygous loci are identified for the reference chromosomal region of the first parental genome. Each reference heterozygous locus includes a corresponding first reference allele in a first reference haplotype and a corresponding second reference allele in a second reference haplotype in the reference chromosomal region of the first parental genome. Referring back to the example of Figure 15, the reference heterozygous locus has a corresponding first target allele of {A,T,C} on Hap I and a corresponding second target of {T,C,A} on Hap II allele. At step 1775, a plurality of free DNA molecules of the reference set are identified. Each episomal DNA molecule of the reference set is located at any of the reference heterozygous loci, includes the corresponding first reference allele, and includes at least one of the K genomic loci in the reference chromosomal region. At step 1780, the degree of methylation of the K reference mixture is measured at the K genomic loci using a reference set of free DNA molecules. At step 1785, the first reference percent contribution of fetal tissue type is determined in the mixture using the K reference methylation degree. At step 1790, a first separation value between the first target percent contribution and the first reference percent contribution is calculated. At step 1795, the first separation value is compared to a threshold value to determine whether the fetus has a classification of sequence imbalance for the target chromosomal region. If the first separation value exceeds a threshold value, sequence imbalance can be identified. Threshold values can be determined as described above, eg, based on separation values seen in a reference set of samples without sequence imbalance and/or a reference set of samples with sequence imbalance. As an example, a classification may be positive, negative, or indeterminate for the sequence imbalance tested. Depending on the type of sequence imbalance, different thresholds can be used. For example, if the sequence imbalance is missing, the first separation value would be expected to be a negative value. In this case, the threshold value may be a negative number, and the comparison may be based on determining that the first threshold value exceeds the threshold value for a larger negative number. If the sequence imbalance test is amplification, it can test whether the separation value is greater than a threshold value. Thus, the threshold value used may depend on the type of sequence imbalance being tested.VII . Deconvolution of Signatures to Identify Diseased Tissue If genomic signatures (eg, specific SNP alleles) are known, embodiments can determine which tissues are the source of such signatures. Since the episomal DNA molecules exhibiting the signature are from the source tissue, the source tissue can be identified from the percent contribution determined using the episomal DNA molecules exhibiting the signature. Thus, free DNA molecules with the signature of the transplanted organ (eg, the signature of the haplotype of the transplanted organ) can be used to monitor changes in the amount of free DNA molecules from the transplanted organ with high sensitivity, for example in view of the DNA in the mixture A high percentage of the contribution will come from transplanted organs. Examples are provided for transplantation to show that the technique is precise. In another example, a tumor's signature can be used to identify the tissue in which the tumor resides.A . organ transplant As an example of organ transplantation, we analyzed plasma from patients receiving liver transplants and patients receiving bone marrow transplants. For each case, donor-specific SNP alleles were identified via genotyping of tissue from patients and donors. For liver transplant recipients, the donor liver is biopsied and the recipient blood cells are sequenced. For bone marrow transplant cases, buccal swabs (recipient genotype) and blood cells (donor genotype) were sequenced. Plasma DNA samples were sequenced after bisulfite conversion. Sequenced DNA fragments carrying donor-specific SNP alleles and at least one CpG site were used for downstream methylation deconvolution analysis. A total of 72 million and 121 million reads were sequenced for patients receiving liver and bone marrow transplants, respectively. For both cases, 38 and 5355 fragments were used for deconvolution analysis, respectively. Organization type liver transplant recipient bone marrow transplant recipient liver 45.4 4.4 lung 0.0 1.5 colon 29.3 6.3 small intestine 0.0 1.8 pancreas 0.0 0.0 adrenal glands 0.0 0.0 esophagus 0.0 0.0 Adipose tissue 0.0 14.8 heart 0.0 0.0 brain 14.5 9.6 T cells 0.0 12.3 B cells 5.9 16.6 Neutrophils 4.9 32.8 surface 6 . Percentage contribution of different organs in two transplant recipients to plasma DNA fragments carrying donor-specific alleles. Table 6 shows methylation deconvolution analysis of plasma DNA fragments carrying donor-specific alleles in liver and bone marrow transplant recipients. Numbers represent the percent contribution of different tissues to donor-specific plasma DNA fragments. For liver transplant cases, the liver was shown to be the most important contributor to these DNA fragments. In the case of bone marrow transplantation, the hematopoietic system (including T cells, B cells, and neutrophils) is the major contribution of donor-specific DNA fragments. These results indicate that methylation deconvolution can accurately indicate the tissue origin of DNA fragments with single nucleotide changes. The small number of sequenced fragments are likely due to measurement inaccuracies attributable to other tissues, since a relatively small number of donor-specific fragments were used for deconvolution analysis. The percent contribution of tissue associated with the transplanted organ can be determined and monitored in the manner described above. Where the baseline percent contribution (an example of reference percent contribution) is relatively high due to the use of only free DNA molecules exhibiting the donor signature, smaller changes in the total amount of donor DNA in plasma can be detected. Therefore, methylation deconvolution analysis can be applied to monitor organ transplantation. As seen above with respect to liver transplantation, methylation deconvolution is not absolutely specific. In this analysis, plasma DNA fragments carrying donor-specific alleles were used for methylation deconvolution analysis. These fragments are specific to the donor and should only be derived from the liver of this liver transplant recipient. Therefore, the theoretical contribution of the liver should be 100%. Another possibility is that certain cell types are present in different types of tissues such that the liver methylation profile overlaps with other tissues. For example, connective tissue cells in the liver may also be present in other organs. However, it is possible to identify whether relative percentages of other samples from other patients or patients of the invention (eg, at other times) release more free DNA molecules. In various embodiments, the donor signature may correspond to a particular haplotype of the donor genome or two haplotypes in a chromosomal region. Methylation deconvolution can be performed using free DNA molecules located on a particular donor haplotype, and the increase in the percentage contribution of a particular haplotype can be monitored. If there is a significant increase (eg, as measured by a percentage or absolute threshold), rejection of the transplanted organ can be identified. 18 shows a graphical representation of haplotype deconvolution for organ transplantation monitoring according to an embodiment of the present invention. The donor has haplotypes labeled Hap I and Hap II, and the recipient has haplotypes labeled Hap III and Hap IV. The donor has signatures at locus 1 and 3, since the allele was not found on the recipient haplotype. Locus 2 and locus 4 do not have a donor signature. Thus, embodiments may use DNA molecules located at locus 1 and locus 3 as part of the deconvolution method. Plasma DNA deconvolution can be used to determine that the measured percentage contribution from the transplanted organ increases at or relative to baseline. In some embodiments, the percentage contribution can be determined separately for each of Hap I and Hap II if different signatures are present; such different signatures can be present at different loci. In other embodiments, a single percent contribution can be determined for both haplotypes, such as when they share a signature. In the example shown in Figure 18, Hap I and Hap II share a signature at locus 1 and locus 3. Thus, the contribution of transplanted organs can be determined using haplotype deconvolution. An increase in the haplotype contribution to the transplanted organ would be appropriate to indicate the organ's contribution to the increase in plasma DNA. In various embodiments, the baseline level can be determined from a population of transplant recipients without rejection or from a population of transplant recipients with rejection. When using recipients with rejection, baseline levels can be determined to be lower than those from a population of transplant recipients with rejection. As mentioned above, the donor can have two identical haplotypes or the recipient can also have two identical haplotypes. In addition, the donor and recipient may share haplotypes. As long as the donor or recipient has a unique haplotype, changes in the percentage of free DNA molecules from the donor tissue can be determined. In the former, rejection will be detected upon seeing an increased contribution from the donor's unique haplotype in the plasma (or other sample). In the latter, rejection will be detected when a reduction in the contribution of receptor unique haplotypes in plasma is seen. Thus, some embodiments may use a first haplotype that is present in normal cells of an organism and not present in abnormal cells that may be in the mixture. This would correspond to the latter example above, when the receptor has a unique haplotype. Another example is when the patient has a unique haplotype in healthy cells compared to the tumor (eg previously found in the organism). In this embodiment, the first tissue type may be determined to have a disease condition when the first separation value is less than a threshold value. In some embodiments, treatment may be provided if the transplanted organ is detected as rejected. For example, variations in dosage of anti-rejection drugs can be provided. As another example, a new organ may be obtained, and surgery may be performed to remove the old transplanted organ and place the new transplanted organ.B . hepatocellular carcinoma ( HCC ) As an example of determining the source tissue of a cancer signature or bias (or monitoring a known or pre-existing tumor), we analyzed plasma from HCC patients. The patient's tumor and blood cells were sequenced to identify cancer-specific single nucleotide mutations. Sequenced DNA fragments carrying cancer-specific mutations and at least one CpG site were used for downstream methylation deconvolution analysis. A total of 11,968 fragments were used for deconvolution analysis. In addition to methylation profiles from normal tissues and organs, we also included methylation profiles of HCC tissues as candidate source tissues. In another embodiment, more types of tumor tissue may be considered candidate tissues for mutation. In one embodiment, methylation profiles of common cancers such as, but not limited to, colorectal, lung, breast, pancreatic, prostate, bladder, cervical, and ovarian cancers can be included as candidate tissues. In another embodiment, only the most likely cancer specific to the patient may be included in the analysis. For example, in female patients, breast, ovarian, colorectal, and cervical cancer are considered. In another embodiment, ethnic origin and age are considered in the selection of candidate tissues. Table 7 shows methylation deconvolution of plasma DNA fragments bearing cancer-associated mutations. Deconvolution analysis accurately determined that DNA fragments harboring cancer-associated mutations were mainly derived from liver cancer tissue. organization Contribution ( % ) liver 0.0 lung 0.0 colon 0.0 small intestine 0.0 pancreas 0.0 adrenal glands 0.0 esophagus 0.0 Adipose tissue 0.0 heart 0.0 brain 0.0 T cells 0.0 B cells 0.0 neutrophils 4.6 liver cancer 95.4 placenta 0.0 surface 7 . Percent contribution of HCC patients using cancer mutations. In some embodiments, tumors can be initially identified by detecting copy number variations, eg, as described in US Pat. Nos. 8,741,811 and 9,121,069. Specific tissue of origin can be determined based on previously identified patterns of copy number variation in various tumors, eg, as described in US Patent Application 14/994,053. Once the tumor has been identified, treatment can be performed, eg, by surgery, radiation therapy, or chemotherapy. In any event, a biopsy can be obtained after assaying the source tissue. Cancer-specific point mutations can be determined from biopsies or from DNA fragments in plasma (eg, as described in US Patent Publication 2014/0100121, or other mixtures associated with copy number variation. Following treatment, key changes will be the disappearance of genomic biases, including copy number variations and point mutations. When these biases disappear, analysis of the genomic signatures of point mutations in the affected regions will yield changes in tissue contribution via methylation deconvolution analysis. If the tumor recovers in the future, cancer-related changes in tissue composition (as determined using methylation deconvolution analysis) will again be visible. For example, the percentage contribution can be compared to a reference percentage contribution, and if a change is detected, a new treatment schedule can be provided. In various embodiments, the cancer-specific mutation may be on only one haplotype or on both haplotypes, eg, in a manner similar to the donor examples above. Thus, if different signatures are present, like a donor, the percent contribution can be determined separately for each of Hap I and Hap II; such different signatures can be present at different loci. In other embodiments, a single percent contribution can be determined for both haplotypes, eg, when they share a signature.C . blot In another embodiment, haplotype deconvolution analysis can be applied to the analysis of genomic regions showing tissue-specific imprints. Differential methylation of paternally and maternally inherited alleles in different tissues and organs has been shown to be normal (Baran et al. Genome Res 2015;25:927-36). Haplotype deconvolution will be suitable for monitoring the contribution of organs exhibiting tissue-specific imprinting. For example, methylation deconvolution can be performed on both paternally and maternally inherited haplotypes when they have different methylation states in the liver but not in other tissues. In one embodiment, both paternal and maternal methylation patterns can be included as candidate tissues in the analysis.D . Methods using genomic signatures 19 is a flowchart illustrating a method 1900 of analyzing a biological sample of an organism to detect whether a first tissue type has a disease condition associated with a first haplotype according to embodiments of the present invention. A biological sample includes a mixture of free DNA molecules from multiple tissue types, including a first tissue type. Method 1900 is performed at least in part using a computer system. At step 1910, a plurality of free DNA molecules from the biological sample are analyzed. Step 1910 may be performed using the techniques described in step 140 of method 100 of FIG. 1 . For example, at least 1,000 free DNA molecules can be analyzed to determine where the free DNA molecules are located, and the degree of methylation can be measured as described below. In addition, the free DNA molecules are analyzed to determine the individual alleles of the free DNA molecules. For example, alleles of DNA molecules can be sequenced reads or specific probes that hybridize to DNA molecules. At step 1920, one or more loci are identified. Each locus has the first allele on the first haplotype of the first chromosomal region. The first haplotype has any of the following properties: (1) not present in healthy cells of the organism, but may alternatively be derived from a tumor or transplanted tissue (as an example); or (2) present in the organism In normal cells and not in abnormal cells that can be in the mixture. Thus, the first haplotype has a genomic signature. In this way, there is a difference between healthy (normal) cells and abnormal cells, allowing embodiments to track the percent contribution of one or the other or both to track the extent (eg, percent contribution) of abnormal cells. In the case of property (1), the first haplotype is associated with a disease condition such as cancer or rejection of transplanted tissue. Thus, a particular cancer may have the first haplotype in the cancer genome of that particular cancer. One or more first classes at one or more loci on a first haplotype can be identified by obtaining a tissue sample (eg, tumor or transplant tissue) and analyzing the DNA molecules of the tissue sample to determine the first haplotype allele. Such tissue samples can be obtained from biopsies, and the method 1900 can be used to test whether the cancer has metastasized to other tissues, or has recurred after surgery. Each of the loci can be a heterozygous locus or a homozygous locus in the abnormal cell. For example, in Figure 18, locus 1 and locus 3 are homozygous in the donor organ. But in the end, more than one allele will be observed in plasma for all loci, as each locus will have a signature for healthy cells or for abnormal cells. Thus, there will be two haplotypes across tissue types, but a single tissue type may have only one haplotype in the area analyzed. At step 1930, a first plurality of free DNA molecules are identified. Each of the plurality of free DNA molecules is located at any of the loci from step 1920 and includes the corresponding first allele at one locus, so that the free DNA molecule can be identified as corresponding to the first a haplotype. Each of the first set of free DNA molecules also includes at least one of the N genomic loci, wherein the genomic locus is used to measure the degree of methylation. N is an integer, eg, greater than or equal to 2, 3, 4, 5, 10, 20, 50, 100, 200, 500, 1,000, 2,000, or 5,000. At step 1940, the methylation level of the N first mixture of N genomic loci is measured using a first set of a plurality of free DNA molecular weights. A first mixture degree of methylation can be measured for each of the N genomic loci. Step 1940 may be performed in a similar manner to step 150 of method 100 of FIG. 1 . In some embodiments, the degree of methylation of DNA molecules can be measured using methylation sensing sequencing results, which can also be used to determine the location and individual alleles of DNA molecules. At step 1950, a first percent contribution of a first tissue type in the mixture is determined using the N first methylation levels. In some embodiments, step 1950 may be performed via steps 160 and 170 of method 100 of FIG. 1 . Thus, the percent contribution can be determined simultaneously for a set of M tissue types. Step 1950 may use the N tissue-specific methylation levels for N genomic loci determined for each of the M tissue types, eg, as in step 120 of method 100 of FIG. 1 . At step 1960, a separation value between the first percent contribution and the reference percent contribution is calculated. Examples of separate values are described herein. The reference percent contribution can be determined using a sample from an organism that is healthy for the first tissue type. For the transplant example, the reference percent contribution can be determined from one or more measurements of a biological sample from an organism in which the transplanted first tissue is not rejected. At step 1970, the separation value may be compared to a threshold value to determine whether the first tissue type has a classification of a disease condition. For example, if the first haplotype is associated with cancer, an appreciable first percentage contribution indicates that the first tissue type has cancer, as can be measured by a separation value above a threshold value (eg, when a reference percentage when the contribution is zero). An amount in which the first percent contribution exceeds a threshold value may be indicative of some degree of cancer. As another example, a first haplotype can be specific for transplanted tissue, and a high contribution relative to a reference can indicate that the organism rejects the transplanted tissue. In one embodiment where the first haplotype is present in normal cells of the organism and not in abnormal cells that may be in the mixture, when the first segregation value is less than a threshold value, the first tissue type can be determined as Have a disease condition. An example of a disease condition is preeclampsia, which can be associated with a spectrum of pathological changes in fetal tissue such as the placenta. For example, in such cases, if the first haplotype is specific to the fetus, eg, a paternally inherited haplotype, it may be increased in maternal plasma of pregnant women with concurrent preeclampsia. In some embodiments, a second haplotype for diseased tissue, such as transplanted tissue or tumors, may also be used. Accordingly, a second percent contribution can be calculated and compared to the reference percent contribution. Thus, a second set of plural episomal DNA molecules can each be located at any of one or more loci, including the corresponding second allele on the second haplotype of the first chromosomal region, and including N at least one of the genomic loci. The second haplotype will have the same properties only from healthy cells or abnormal cells. Multiple tissue types can be tested (eg, using method 100 of FIG. 1 ) to determine the tissue of origin of the first haplotype, eg, when it is associated with cancer. Thus, the N first methylation levels can be used to determine the percent contribution of other tissue types in the mixture, and the corresponding separation value between the corresponding percent contribution and the respective reference percent contribution can be compared to a threshold value to determine the other tissue A classification of whether each of the types has a specific cancer. Different tissues may have different reference percentage contributions.VIII. identify cancer CNAs source organization In some embodiments, the origin of the tumor may not be known. Thus, it may be difficult to identify point mutations in tumors, as may be used in method 1900 of Figure 19 or other methods described herein. Additionally, tumors may not have numerous point mutations, but may have chromosomal regions that exhibit amplifications and deletions (examples of copy number variation). To address this issue, embodiments may use replica number analysis to identify regions exhibiting copy number variation (CNAs). Typically, CNAs appear on only one haplotype of a region. Since only one haplotype has an amplification or deletion, there will be relatively large differences between the percentage contributions of the tissue types in which the tumor resides. CNA analysis can be performed in a variety of ways, eg, as described in US Pat. Nos. 8,741,811 and 9,121,069. For example, the human genome (or that of other types of organisms) can be partitioned into approximately 3,000 non-overlapping 1 Mb partitions. The number of reads mapped to each 1 Mb partition can be determined. After correcting for GC bias (Chen EZ et al. (2011)PLoS One 6(7):e21791), the sequence read density for each partition can be calculated. For each partition, the sequence read density of the test cases can be compared to the value of the reference control individual. Increases or decreases in the number of replicates can be defined as 3 standard deviations above and below the mean of the control, respectively. Thus, a first chromosomal region can be identified as exhibiting copy number variation based on a first amount of free DNA molecules located in the first chromosomal region. To determine the tissue origin of copy number variation in plasma, plasma DNA tissue mapping can be performed using methylation markers located within genomic regions exhibiting such deviations in plasma. In the following example for a cancer patient, the mapping of plasma DNA copy number variation was performed only if the bias affected a contiguous chromosomal region of at least 30 Mb to make a sufficient number of methylation markers available for the mapping.A . Identification with copy number variation ( CNAs ) area of A 62-year-old male patient with HCC was recruited with informed consent from the Department of Surgery, Prince of Wales Hospital, Hong Kong. Ten milliliters of venous blood were collected in EDTA tubes at diagnosis and 3 months after tumor resection. Blood samples were centrifuged at 3000 g for 10 minutes to separate blood cells from plasma. Plasma was centrifuged at 30,000 g for an additional 10 minutes to remove remaining cells. DNA extracted from blood cells was used to phase SNPs to construct the patient's haplotype using the 10x genomics platform following the manufacturer's instructions. High molecular weight DNA was extracted from blood or tissue samples using the MagAttract HMW DNA kit (QIagen, Germany). DNA quality was verified by genomic DNA analysis ScreenTape on a 4200 TapeStation system (Agilent, Germany). DNA was quantified by the dsDNA HS Analysis Kit on a Qubit 3.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA). Sample indexing and library preparation were performed using the GemCode system and its associated reagents (1OX Genomics, Pleasanton, CA) (Zheng et al. Nat Biotechnol. 2016 Mar;34:303-11). Briefly, 1 ng of DNA was input for a GEM reaction, where individual DNA molecules were fragmented to introduce specific barcodes and extend the DNA. After the GEM reaction, a sequenced library was prepared according to the manufacturer's recommendations. Libraries were quantified by qPCR using the KAPA library quantification kit (KAPA Biosystems, Wilmington, MA). Normalized libraries were sequenced on a HiSeq 2500 sequencer (Illumina, San Diego, CA) by paired-end sequencing of 98 bp, 14 bp 15 and 8 bp 17 indexed reads. Sequencing results were analyzed using the Long Ranger software suite (10X Genomics) such that all heterozygous SNPs were phased and the patients were determined for both haplotypes. Plasma samples were sequenced to a depth of 17× using Illumina. Copy number variation was detected in the plasma of HCC patients according to methods as previously described (Chan et al. Clin Chem. 2013;59:211-24). Figure 20 shows a graph of copy number variations detected in the plasma of HCC patients according to embodiments of the present invention. Inner circles represent results from plasma samples collected at diagnosis (pre-operation) and outer circles represent results from plasma samples collected 3 months after tumor resection (post-operation). Each point represents a 1 Mb region. The green, red, and gray dots represent the areas of replica count gain, replica count loss, and no replica count change, respectively. Copy number variations were detected in plasma samples at diagnosis and these changes disappeared after tumor removal. In Figure 20, two regions are highlighted as having CNAs. Region 2010 has a replica count increase, and region 2020 has a replica count loss. The haplotypes of these regions can be determined using any tissue sample from an individual, and not just tumor samples. The difference in the number of replicates is what drives the difference in the percentage contribution, and the difference should be greatest in the tissue types with tumors.B . Determination of tissue sources of copy number variation We performed methylation deconvolution analysis independently for the two haplotypes. For illustrative purposes, the two haplotypes are named Hap I and Hap II. Plasma DNA molecules covering the hybrid SNP and at least one CpG site were used for this analysis. Plasma DNA molecules carrying the SNP allele on Hap I were analyzed independently from those carrying the allele on Hap II. The methylation status of CpG sites was used for methylation deconvolution to map molecules to Hap I and Hap II independently. Thus, the tissue contribution to Hap I and Hap II in plasma DNA can be determined. First, we focus on regions with amplification. For illustrative purposes, we analyzed the amplified region on chromosome 1q (as an example). At the time of diagnosis after tumor resection Hap I 34,119 11,131 Hap II 26,582 11,176 Table 11 shows the number of sequence reads from the two haplotypes. At diagnosis, the number of reads mapped to Hap I increased compared to the number of reads mapped to Hap II. This indicates that Hap I is amplified relative to Hap II. This observation is compatible with the fact that certain chromosomes are replicated in cancer rather than both homologous chromosomes that are amplified to the same extent, and is consistent with the fact that copy number variation occurs preferentially in one haplotype (Adey A. et al., Nature. 2013;500:207-11; LaFramboise T. et al, PLoS Comput Biol. 2005;1(6):e65). The dose difference between the two haplotypes disappeared after tumor resection. The difference in the absolute number of sequence reads between the plasma samples obtained at diagnosis and after tumor resection is due to the difference in the total number of sequence reads generated for the two plasma samples. At the time of diagnosis after tumor resection Hap I Hap II difference Hap I Hap II difference liver 19.7 8.0 11.7 21.3 21.9 -0.6 lung 5.4 0 5.4 0 0 0 colon 0 0 0 0 0 0 brain 0 0 0 9.0 9.0 0 heart 0 17.0 -17 3.0 2.5 -0.5 blood cells 74.9 75.0 0 66.7 66.6 0.1 total 100 100 0 100 100 0 Table 12 shows the percent contribution of different tissues to plasma DNA for the two haplotypes at diagnosis and after tumor resection. At diagnosis, the liver contribution to plasma DNA was 19.7% and 8.0% for Hap I and Hap II, respectively. The highest difference between different types of organizations was 11.7%. This indicated that the dose difference between Hap I and Hap II in plasma most likely contributed from the liver contribution. This additionally indicates that a possible source of chromosomal aberrations is from the liver, as the variation in the number of copies is most likely attributable to the replication of Hap I in the analysis of sequence read counts. In another example, the difference in the contributions of Hap I and Hap II can be ranked to indicate the relative likelihood that different tissues are the origin of the copy number variation. The value for heart is -17, which is in the opposite direction to the copy number variation identified by Table 11. Thus, although the absolute value for the heart is greater than that for the liver, the opposite sign discounts the heart as a viable candidate for the tissue type of origin of the tumor. Since the total contribution of all organs is 100%, a positive difference in the contribution of the liver causes other tissues to have negative values. Similarly, this haplotype-specific methylation deconvolution can also be performed on regions with loss of replica number. For illustrative purposes, we performed this analysis on a region on chromosome lp that exhibits loss of duplicate number. At the time of diagnosis after tumor resection Hap I 19,973 8,323 Hap II 12,383 7,724 Table 13 shows the number of sequence reads from the two haplotypes. At diagnosis, the number of reads mapped to Hap II is reduced compared to the number of reads mapped to Hap I. In tumor tissue, the majority of regions with loss of chromosome copy number will involve only the deletion of one of the two chromosomes. Therefore, the relative reduction in the dose of Hap II is compatible with the absence of Hap II. The difference in dose of the two haplotypes that disappeared after tumor resection indicates that the amount of tumor-derived DNA has decreased or disappeared from plasma. At the time of diagnosis after tumor resection Hap I Hap II Difference (Hap I-Hap II) Hap I Hap II Difference (Hap I-Hap II) liver 13.3 5.5 7.8 10.2 13.2 -3 lung 0 0 0 4.1 0.5 3.6 colon 3.8 0 3.8 8.6 17.5 -8.9 brain 0 0 0 0 0 0 heart 3.7 0 3.7 25.5 19.4 6.1 blood cells 79.2 94.5 -15.3 51.6 49.4 2.2 total 100 100 0 100 100 0 Table 14 shows the percent contribution of different tissues to plasma DNA for the two haplotypes at diagnosis and after tumor resection. At diagnosis, the liver contribution to plasma DNA was 13.3% and 5.5% for Hap I and Hap II, respectively. The highest difference between different types of organizations was 7.8%. This indicated that the dose difference between Hap I and Hap II in plasma most likely contributed from the liver contribution. This additionally indicates that the likely source of chromosomal aberrations is from the liver, as the variation in the number of copies is most likely due to the deletion of Hap II in the analysis of sequence read counts. In another example, the difference in the contributions of Hap I and Hap II can be ranked to indicate the relative likelihood that different tissues are the origin of the copy number variation.C . Methods of Determining Tissue Origin of Tumors 21 is a flow chart illustrating a method of analyzing a biological sample of an organism to identify the origin of chromosomal aberrations in accordance with embodiments of the present invention. A biological sample includes a mixture of free DNA molecules from a plurality of tissue types including the first tissue type. At step 2110, a plurality of free DNA molecules from the biological sample are analyzed. Step 2110 may be performed using the techniques described in step 1910 of FIG. 1 and step 140 of method 100 of FIG. 1, as well as other steps describing similar features. At step 2115, the first chromosomal region is identified as exhibiting copy number variation in the organism based on the first amount of free DNA molecules located in the first chromosomal region. For example, plasma DNA analysis is performed to identify regions exhibiting copy number variation. This distortion may correspond to over- or under-representation. In some embodiments, the genome can be divided into partitions (eg, 1 Mb partitions), and the amount of free DNA molecules from a particular partition can be determined (eg, by mapping sequence reads to that portion of the reference genome). The amount of a particular partition can be normalized (eg, with respect to the average amount of partitions), and over- or under-representation can be identified. Techniques other than counting the mapping of DNA molecules to specific regions can be used. For example, the size distribution of DNA molecules aligned with a first chromosomal region can be used to detect CNAs. For example, cell-free tumor DNA is smaller than cell-free DNA from normal cells. This size difference can be used to detect differences in size distribution (eg, average size or ratio of the number of DNA molecules of different sizes) between two haplotypes of the region, or between the region and another region. At step 2120, the first and second haplotypes of the organism in the first chromosomal region are determined. Both haplotypes can be determined as part of step 2115. The two haplotypes can be determined using the same free mixture or from different samples, such as cell samples. At step 2130, one or more heterozygous loci of the first chromosomal region are identified. Each heterozygous locus includes a corresponding first allele in the first haplotype and a corresponding second allele in the second haplotype. Step 2130 may be performed in a similar manner to other similar steps of the methods described herein. At step 2140, a first plurality of free DNA molecules are identified. Each DNA molecule of the first set is located at any one of one or more heterozygous loci, including the corresponding first allele of the heterozygous locus, and including at least one of the N genomic loci. N is an integer greater than or equal to 2. Step 2140 may be performed in a manner similar to other similar steps of the methods described herein. At step 2150, the methylation levels of the N first mixtures of N genomic loci are measured using a first set of a plurality of free DNA molecular weights. Step 2150 may be performed in a manner similar to other similar steps of the methods described herein. At step 2160, a second plurality of free DNA molecules are identified. Each DNA molecule of the second set is located at any of one or more heterozygous loci, including the corresponding second allele of the heterozygous locus, and including at least one of the N genomic loci. Step 2160 may be performed in a similar manner to other similar steps of the methods described herein. In some embodiments, a first number of free DNA molecules in a first plurality of free DNA molecules can be determined, and a second number of free DNA molecules in a second plurality of free DNA molecules can be determined , for example as shown in Table 11. Which number is higher can be determined, thereby providing the source tissue with information about the expected segregation value, such as which haplotype should have a higher percentage contribution. The first plurality of free DNA molecules can have a first size distribution, and the second plurality of free DNA molecules can have a second size distribution. A statistical value of the size distribution of the DNA molecules can be determined for each haplotype, thereby providing a first statistical value and a second statistical value. Haplotypes with a smaller size distribution would be expected to have a higher number of copies than other haplotypes because tumor cell-free DNA is known to be smaller, as described in US Pat. No. 8,741,811. Examples of statistical values for size distribution are the ratio of numbers of DNA molecules of different sizes, the average size, or the percentage of DNA molecules of a particular size (eg, below a size threshold). At step 2170, the N second mixture of N genomic loci is measured for methylation degree using a second plurality of free DNA molecular weights. Step 2170 may be performed in a similar manner to other similar steps of the methods described herein. Steps 2180 and 2190 may be performed for each of a plurality of M tissue types. The M tissue types can include a predetermined list of tissue types that are screened and for which a reference methylation level can be known. A preset list may include the most prominently visible cancer tissue. M is an integer greater than 1. At step 2180, the computer system uses the N first methylation levels to determine the corresponding first percent contribution of the tissue types in the mixture. The computer system uses the N second degrees of methylation to determine the corresponding second percent contribution of the tissue types in the mixture. Step 2180 may be performed in a similar manner to other similar steps of the methods described herein. At step 2190, a corresponding separation value between the corresponding first percent contribution and the corresponding second percent contribution is calculated. Various separation values can be used, eg, as described herein. At step 2195, the first tissue type is identified as the origin of the copy number variation based on the first segregation value of the first tissue type having the largest among the corresponding segregation values. The assay may require the highest separation value to be sufficiently higher than the second highest separation value. For example, the difference may be required to be at least a threshold value, such as 1%, 2%, 3%, 4%, 5%, 6% or 7%. In one embodiment, the difference between the first segregation value and the next highest segregation value may be compared to a threshold value to determine a classification of the degree of likelihood that the first tissue type is the origin of the copy number variation. Therefore, even if the difference is not above a threshold value, a probability or other classification can be provided. For example, a linear relationship from 0 to a threshold value can be used, where once the difference is equal to the threshold value, the probability is 100%. Depending on how the separation value is determined, the maximum value can be either the largest negative number or the largest positive number. For example, the differences in Table 14 can be determined using Hap II-Hap I. The maximum value can be determined to be positive or negative using analysis of DNA molecules on each haplotype, such as counts as in Table 13 or size analysis as described above. In some embodiments, the segregation value can always be determined such that a maximum positive value is expected, eg, by subtracting the percent contribution of haplotypes with a lower number of copies from the percent contribution of haplotypes with a higher number of copies. After identification of the source, studies using imaging modalities such as computed tomography (CT) scans or magnetic resonance imaging (MRI) of the individual (either the entire individual or specifically, the candidate organ) can be performed to confirm or rule out the tumor in the organ existence. If a tumor is confirmed, treatment, such as surgery (with a scalpel or with radiation) or chemotherapy, may be administered.IX . computer system Any computer system mentioned herein may utilize any suitable number of subsystems. An example of such a subsystem is shown in the computer apparatus 10 of FIG. 22 . In some embodiments, the computer system includes a single computer device, wherein the subsystems may be components of the computer device. In other embodiments, a computer system may include a plurality of computer devices with internal components, each of which is a subsystem. Computer systems may include desktop and laptop computers, tablet computers, mobile phones, and other mobile devices. The subsystems shown in FIG. 22 are interconnected via system bus 75 . Additional subsystems are shown, such as printer 74, keyboard 78, storage device 79, monitor 76 coupled to display adapter 82, and the like. Peripherals and I/O devices coupled to input/output (I/O) controller 71 may be provided by any number of means known in the art, such as input/output (I/O) port 77 (eg, USB , FireWire® )) is connected to the computer system. For example, I/O port 77 or external interface 81 (eg, Ethernet, Wi-Fi, etc.) may be used to connect computer system 10 to a wide area network (eg, the Internet, a mouse input device, or a scanner) . Interconnection via system bus 75 allows central processing unit 73 to communicate with the various subsystems and control the execution of a plurality of instructions from system memory 72 or storage device 79 (eg, a hard disk, such as a hard drive, or an optical disk). execution, and the exchange of information between subsystems. System memory 72 and/or storage device 79 may embody computer-readable media. Another subsystem is data acquisition devices 85, such as cameras, microphones, accelerometers, and the like. Any data mentioned herein can be output from one component to another and can be output to a user. A computer system may include a plurality of identical components or subsystems connected together, for example, by external interface 81 or by internal interface. In some embodiments, computer systems, subsystems, or devices may communicate via a network. In such cases, one computer may be considered a client and the other computer may be considered a server, each of which may be part of the same computer system. Each of the client and server may include multiple systems, subsystems or components. Aspects of the embodiments may be implemented using hardware (such as application specific integrated circuits or field programmable gate arrays), in logic control and/or using computer software, using ordinary programmable processors, in modular or Implemented in an integrated manner. As used herein, a processor includes a single-core processor on the same integrated die, a multi-core processor, or multiple processing units on a single circuit board or networked. Based on the disclosure and teachings provided herein, those of ordinary skill in the art will know and appreciate other ways and/or methods of implementing embodiments of the invention using hardware and combinations of hardware and software. Any software components or functions described in this application are available as to be executed by the processor using any suitable computer language (such as Java, C, C++, C#, Objective-C, Swift) or scripting language (such as Perl or Python) ), executed using, for example, conventional or object-oriented techniques. The software code may be stored on a computer-readable medium in the form of a series of instructions or commands for storage and/or transmission. Suitable non-transitory computer-readable media may include random access memory (RAM), read only memory (ROM), magnetic media such as a hard disk drive or a floppy disk drive, or optical media such as a compact disc (CD) Or DVD (Digital Versatile Disc), flash memory, and the like. The computer-readable medium can be any combination of these storage or transmission devices. The programs may also be encoded and transmitted using carrier signals suitable for transmission over wired, optical, and/or wireless networks conforming to various protocols, including the Internet. Thus, computer-readable media can be created using data signals encoded by such programs. Computer-readable media encoded with code may be packaged with compatible devices or provided separately from other devices (eg, via Internet download). Any such computer-readable media can exist on or within a single computer product (eg, a hard drive, a CD, or an entire computer system), and can exist on or within different computer products within a system or network. The computer system may include a monitor, printer, or other suitable display for providing any of the results mentioned herein to the user. Any of the methods described herein can be performed, in whole or in part, using a computer system that includes one or more processors that can be configured to perform the steps. Thus, embodiments may be directed to computer systems configured to perform the steps of any of the methods described herein, potentially using different components to perform individual steps or groups of steps. Although the steps of the methods herein are presented as numbered steps, they can be performed simultaneously or in a different order. Additionally, portions of these steps may be used with portions of other steps of other methods. In addition, all or part of the steps may be optional. Additionally, any steps in any method may be performed using modules, units, circuits, or other means for performing such steps. The specific details of a particular embodiment may be combined in any suitable manner without departing from the spirit and scope of the embodiments of the present invention. However, other embodiments of the invention may be directed to specific embodiments associated with each individual aspect or a specific combination of such individual aspects. The foregoing description of example embodiments of the present invention has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form described, and many modifications and variations are possible in light of the above teachings. The statement "a/an" or "the" is intended to mean "one or more" unless specifically indicated to the contrary. Unless expressly indicated to the contrary, the use of "or" is intended to mean "inclusive or" and not "exclusive or". Reference to a "first" component does not necessarily require the provision of a second component. Furthermore, unless expressly stated otherwise, reference to a "first" or "second" element does not limit the reference to a particular location of the element. All patents, patent applications, publications, and descriptions mentioned herein are incorporated by reference in their entirety for all purposes. Neither is allowed to be prior art.

10:電腦系統 71:輸入/輸出控制器 72:控制系統記憶體 73:中央處理器 74:印表機 75:系統匯流排 76:監測器 77:輸入/輸出埠 78:鍵盤 79:儲存裝置 81:外部介面 82:顯示配接器 100:方法 110:步驟 120:步驟 130:步驟 140:步驟 150:步驟 160:步驟 170:步驟 205:生物樣品 210:全基因組亞硫酸氫鹽定序 220:組織特異性甲基化圖譜 230:血漿DNA組織映射 241:產前測試 242:癌症偵測及監測 243:器官移植監測 244:器官損傷評估 300:圖 350:圖 400:表 600:表 700:表 800:表 900:圖 950:圖 1000:圖 1050:圖 1300:方法 1310:步驟 1320:步驟 1330:步驟 1340:步驟 1350:步驟 1360:步驟 1370:步驟 1380:步驟 1385:步驟 1390:步驟 1395:步驟 1400:方法 1410:步驟 1420:步驟 1430:步驟 1440:步驟 1450:步驟 1460:步驟 1470:步驟 1480:步驟 1700:方法 1710:步驟 1720:步驟 1730:步驟 1740:步驟 1750:步驟 1760:步驟 1770:步驟 1780:步驟 1785:步驟 1790:步驟 1795:步驟 1900:方法 1910:步驟 1920:步驟 1930:步驟 1940:步驟 1950:步驟 1960:步驟 1970:步驟 2010:區域 2020:區域 2100:方法 2110:步驟 2115:步驟 2120:步驟 2130:步驟 2140:步驟 2150:步驟 2160:步驟 2170:步驟 2180:步驟 2190:步驟 2195:步驟10: Computer System 71: Input/Output Controller 72: Control system memory 73: CPU 74: Printer 75: System busbar 76: Monitor 77: input/output port 78: Keyboard 79: Storage Device 81: External interface 82: Display adapter 100: Method 110: Steps 120: Steps 130: Steps 140: Steps 150: Steps 160: Steps 170: Steps 205: Biological Samples 210: Whole Genome Bisulfite Sequencing 220: Tissue-specific methylation profiles 230: Plasma DNA tissue mapping 241: Prenatal Testing 242: Cancer Detection and Surveillance 243: Organ Transplant Monitoring 244: Organ Damage Assessment 300: Figure 350: Figure 400: table 600: table 700: table 800: table 900: Figure 950: Figure 1000: Figure 1050: Figure 1300: Method 1310: Steps 1320: Steps 1330: Steps 1340: Steps 1350: Steps 1360: Steps 1370: Steps 1380: Steps 1385: Steps 1390: Steps 1395: Steps 1400: Method 1410: Steps 1420: Steps 1430: Steps 1440: Steps 1450: Steps 1460: Steps 1470: Steps 1480: Steps 1700: Method 1710: Steps 1720: Steps 1730: Steps 1740: Steps 1750: Steps 1760: Steps 1770: Steps 1780: Steps 1785: Steps 1790: Steps 1795: Steps 1900: Method 1910: Steps 1920: Steps 1930: Steps 1940: Steps 1950: Steps 1960: Steps 1970: Steps 2010: Regional 2020: Regional 2100: Methods 2110: Steps 2115: Steps 2120: Steps 2130: Steps 2140: Steps 2150: Steps 2160: Steps 2170: Steps 2180: Steps 2190: Steps 2195: Steps

圖1為說明分析游離之DNA分子之DNA混合物以自根據本發明之實施例之甲基化程度測定來自各種組織類型之百分比貢獻之方法的流程圖。 圖2顯示顯示DNA甲基化解卷積(例如使用血漿)之若干潛在應用及其根據本發明之實施例之應用的示意圖。 圖3A顯示根據本發明之實施例之就15個懷孕女性而言之不同器官對血漿DNA之百分比貢獻的圖。圖3B顯示推論自血漿DNA甲基化解卷積之胎盤貢獻之血漿DNA分率與使用使用根據本發明之實施例的胎兒特異性SNP等位等位基因推論之胎兒DNA百分比濃度之間的相關性之圖350。 圖4顯示根據本發明之實施例測定自懷孕女性中之血漿DNA組織映射分析之百分比貢獻的表。 圖5顯示根據血漿DNA組織映射的除胎盤以外之器官之百分比貢獻及基於根據本發明之實施例的胎兒特異性SNP等位基因之胎兒DNA百分比濃度之圖。 圖6顯示來自根據本發明之實施例之非懷孕健康對照個體組之血漿DNA組織映射分析之百分比貢獻的表。 圖7顯示關於根據本發明之實施例使用第一組標記物(具有高器官特異性)之11個懷孕女性及4個非懷孕健康個體的不同器官對血漿DNA之估計貢獻之表。 圖8顯示關於根據本發明之實施例使用第二組標記物(具有低器官特異性)之11個懷孕女性及4個非懷孕健康個體的不同器官對血漿DNA之估計貢獻之表。 圖9A為顯示估計胎兒DNA百分比濃度(貢獻自胎盤)與藉由計數母體血漿樣品中之胎兒特異性等位基因測定之胎兒DNA百分比濃度之間的相關性之圖。 圖9B為顯示來自甲基化標記物之評估與藉由胎兒特異性等位基因計數測定之胎兒DNA百分比濃度之間的絕對差之圖。 圖10A為顯示根據本發明之實施例使用具有不同選擇標準之標記物推論的胎盤對血漿DNA之貢獻之圖。圖10B為顯示使用在相同組織類型中具有低變化性(類別i)及高變化性(類別ii)之標記物之血漿DNA解卷積之精確性的圖。 圖11A顯示胎兒遺傳來自母親之M等位基因且在根據本發明之實施例之特定基因座處具有MN之基因型的第一情形。圖11B顯示胎兒遺傳來自母親之N等位基因且在根據本發明之實施例之特定基因座處具有NN之基因型的第二情形。 圖12A顯示根據本發明之實施例使用甲基化解卷積由胎兒遺傳之母體單倍型之測定。圖12B顯示根據本發明之實施例之父體單倍型甲基化分析之說明。 圖13為說明使用根據本發明之實施例之甲基化解卷積測定來自母體樣品之胎兒基因組之一部分之方法1300的流程圖。 圖14為說明使用根據本發明之實施例之甲基化程度測定來自母體樣品之胎兒基因組之一部分之方法1400的流程圖。 圖15顯示對於根據本發明之實施例之母體單倍型基於單倍型解卷積之染色體非整倍性偵測。 圖16顯示對於根據本發明之實施例之父體單倍型基於單倍型解卷積之染色體非整倍性偵測。 圖17為根據本發明之實施例使用來自懷孕女性之生物樣品偵測懷孕女性之未出生胎兒之胎兒基因組之一部分中之序列不平衡之方法1700的流程圖。 圖18顯示根據本發明之實施例用於器官移植監測之單倍型解卷積之圖示。 圖19為說明分析生物體之生物樣品以偵測第一組織類型是否具有與根據本發明之實施例之第一單倍型相關之疾病病況之方法的流程圖。 圖20顯示根據本發明之實施例在HCC患者之血漿中偵測之拷貝數變異之圖。 圖21為說明分析生物體之生物樣品以鑑別根據本發明之實施例之染色體畸變之起源之方法的流程圖。 圖22顯示可與根據本發明之實施例的系統及方法一起使用的實例電腦系統10的方塊圖。Figure 1 is a flow chart illustrating a method of analyzing a DNA mixture of free DNA molecules to determine the percentage contribution from various tissue types from methylation levels according to embodiments of the present invention. Figure 2 shows a schematic diagram showing several potential applications of DNA methylation deconvolution (eg using plasma) and its application according to embodiments of the present invention. Figure 3A shows a graph of the percent contribution of different organs to plasma DNA for 15 pregnant women, according to an embodiment of the present invention. 3B shows the correlation between the plasma DNA fraction inferred from the placental contribution of plasma DNA methylation deconvolution and the percent fetal DNA concentration inferred using alleles using fetal-specific SNPs according to embodiments of the present invention Figure 350. Figure 4 shows a table determining the percent contribution from tissue mapping analysis of plasma DNA in pregnant women, according to embodiments of the present invention. 5 shows a graph of the percent contribution of organs other than the placenta mapped from plasma DNA tissue and the percent fetal DNA concentration based on fetal-specific SNP alleles according to embodiments of the present invention. Figure 6 shows a table of percentage contributions from tissue mapping analysis of plasma DNA for a group of non-pregnant healthy control individuals according to embodiments of the present invention. Figure 7 shows a table of estimated contributions to plasma DNA from different organs of 11 pregnant females and 4 non-pregnant healthy individuals using the first set of markers (with high organ specificity) according to embodiments of the present invention. Figure 8 shows a table of estimated contributions to plasma DNA from different organs of 11 pregnant females and 4 non-pregnant healthy individuals using a second set of markers (with low organ specificity) according to embodiments of the present invention. 9A is a graph showing the correlation between estimated fetal DNA percent concentration (contributed from the placenta) and fetal DNA percent concentration determined by counting fetal-specific alleles in maternal plasma samples. Figure 9B is a graph showing the absolute difference between the assessment from methylation markers and the percent fetal DNA concentration determined by fetal-specific allele counting. Figure 10A is a graph showing the inferred contribution of placenta to plasma DNA using markers with different selection criteria according to embodiments of the present invention. Figure 10B is a graph showing the accuracy of plasma DNA deconvolution using markers with low variability (class i) and high variability (class ii) in the same tissue type. Figure 11A shows a first scenario in which a fetus inherits the M allele from the mother and has the genotype of MN at a specific locus according to an embodiment of the invention. Figure 11B shows a second scenario where the fetus inherits the N allele from the mother and has the genotype of NN at a particular locus according to an embodiment of the invention. Figure 12A shows the determination of maternal haplotypes inherited from the fetus using methylation deconvolution in accordance with embodiments of the present invention. Figure 12B shows an illustration of paternal haplotype methylation analysis according to embodiments of the present invention. 13 is a flowchart illustrating a method 1300 for determining a portion of a fetal genome from a maternal sample using methylation deconvolution in accordance with embodiments of the present invention. 14 is a flowchart illustrating a method 1400 for determining a portion of a fetal genome from a maternal sample using methylation levels according to embodiments of the present invention. 15 shows chromosomal aneuploidy detection based on haplotype deconvolution for maternal haplotypes according to embodiments of the present invention. 16 shows chromosomal aneuploidy detection based on haplotype deconvolution for paternal haplotypes according to embodiments of the present invention. 17 is a flowchart of a method 1700 for detecting sequence imbalances in a portion of the fetal genome of a pregnant woman's unborn fetus using a biological sample from a pregnant woman, according to an embodiment of the invention. 18 shows a graphical representation of haplotype deconvolution for organ transplantation monitoring according to an embodiment of the present invention. 19 is a flowchart illustrating a method of analyzing a biological sample of an organism to detect whether a first tissue type has a disease condition associated with a first haplotype according to embodiments of the present invention. Figure 20 shows a graph of copy number variations detected in the plasma of HCC patients according to embodiments of the present invention. 21 is a flow chart illustrating a method of analyzing a biological sample of an organism to identify the origin of chromosomal aberrations in accordance with embodiments of the present invention. 22 shows a block diagram of an example computer system 10 that may be used with systems and methods according to embodiments of the present invention.

100:方法 100: Method

110:步驟 110: Steps

120:步驟 120: Steps

130:步驟 130: Steps

140:步驟 140: Steps

150:步驟 150: Steps

160:步驟 160: Steps

170:步驟 170: Steps

Claims (65)

一種使用來自懷孕女性之生物樣品測定該懷孕女性之未出生胎兒之胎兒基因組之一部分的方法，其中該生物樣品包括來自複數個組織類型，包括母體組織類型及胎兒組織類型之游離(cell-free)之DNA分子之混合物，該胎兒具有父親及母親為該懷孕女性，該方法包含： 藉由電腦系統分析來自該生物樣品之複數個游離之DNA分子，該複數個游離之DNA分子為至少1,000個游離之DNA分子，其中分析游離之DNA分子包括： 鑑別該游離之DNA分子於參考人類基因組中之位置；及 測定該游離之DNA分子之各別等位基因； 測定該胎兒之第一親體的第一親體基因組之第一染色體區域之第一單倍型及第二單倍型； 鑑別該第一親體基因組之第一染色體區域之一或多個雜合基因座，各雜合基因座包括該第一單倍型中之對應第一等位基因及該第二單倍型中之對應第二等位基因； 鑑別第一組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該雜合基因座之對應第一等位基因，且 包括N個基因組位點中之至少一者，N為大於或等於2之整數； 使用該第一組複數個游離之DNA分子量測該N個基因組位點之N個第一混合物甲基化程度； 藉由該電腦系統使用該N個第一甲基化程度測定該混合物中之該胎兒組織類型之第一百分比貢獻(fractional contribution)； 鑑別第二組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該對應第二等位基因，且 包括該N個基因組位點中之至少一者； 使用該第二組複數個游離之DNA分子量測該N個基因組位點之N個第二混合物甲基化程度； 藉由該電腦系統使用該N個第二甲基化程度測定該混合物中之該胎兒組織類型之第二百分比貢獻； 計算該第一百分比貢獻與該第二百分比貢獻之間的第一分離值(separation value)； 基於該第一分離值測定該一或多個雜合基因座之胎兒基因組之部分。A method of determining a portion of the fetal genome of an unborn fetus of a pregnant woman using a biological sample from a pregnant woman, wherein the biological sample includes cell-free from a plurality of tissue types, including maternal and fetal tissue types The mixture of DNA molecules, the fetus has the father and the mother is the pregnant female, the method comprises: Analyzing a plurality of free DNA molecules from the biological sample by a computer system, the plurality of free DNA molecules is at least 1,000 free DNA molecules, wherein analyzing the free DNA molecules includes: identifying the location of the episomal DNA molecule in the reference human genome; and determining the individual alleles of the free DNA molecule; determining the first haplotype and the second haplotype of the first chromosomal region of the first parent genome of the first parent of the fetus; Identifying one or more heterozygous loci in the first chromosomal region of the first parental genome, each heterozygous locus comprising the corresponding first allele in the first haplotype and the second haplotype corresponding to the second allele; A first set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding first allele of the heterozygous locus, and including at least one of N genomic loci, where N is an integer greater than or equal to 2; measuring the methylation degree of the N first mixtures of the N genomic loci using the first set of free DNA molecular weights; Determining, by the computer system, a first fractional contribution of the fetal tissue type in the mixture using the N first methylation levels; A second set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding second allele, and including at least one of the N genomic loci; measuring the degree of methylation of the N second mixtures of the N genomic loci using the second set of free DNA molecular weights; determining, by the computer system, the second percent contribution of the fetal tissue type in the mixture using the N second methylation levels; calculating a first separation value between the first percent contribution and the second percent contribution; The portion of the fetal genome of the one or more heterozygous loci is determined based on the first segregation value. 如請求項1之方法，其中該一或多個雜合基因座為第一複數個雜合基因座。The method of claim 1, wherein the one or more heterozygous loci are the first plurality of heterozygous loci. 如請求項2之方法，其中該第一單倍型為第一母體單倍型，且其中該第二單倍型為第二母體單倍型。The method of claim 2, wherein the first haplotype is a first maternal haplotype, and wherein the second haplotype is a second maternal haplotype. 如請求項3之方法，其進一步包含： 鑑別父體基因組中之第一染色體區域之第二複數個雜合基因座，該第二複數個雜合基因座中之每一者包括第一父體單倍型中之對應第三等位基因及第二父體單倍型中之對應第四等位基因，其中該父體基因組對應於該胎兒之父親； 鑑別第三組複數個游離之DNA分子，其各自： 位於該第二複數個雜合基因座中的任一者， 包括該雜合基因座之該對應第三等位基因，且 包括K個基因組位點中之至少一者； 使用該第三組之第二複數個游離之DNA分子量測該K個基因組位點之K個第三混合物甲基化程度； 使用該K個第三甲基化程度測定該混合物中之該胎兒組織類型之第三百分比貢獻； 鑑別第四組複數個游離之DNA分子，其各自： 位於該第二複數個雜合基因座中的任一者， 包括該雜合基因座之該對應第四等位基因，且 包括該K個基因組位點中之至少一者； 使用該第四組之第二複數個游離之DNA分子量測該K個基因組位點之K個第四混合物甲基化程度； 使用該K個第四甲基化程度測定該混合物中之該胎兒組織類型之第四百分比貢獻； 計算該第三百分比貢獻與該第四百分比貢獻之間的第二分離值； 基於該第二分離值測定該第二複數個雜合基因座之胎兒基因組之部分。The method of claim 3, further comprising: identifying a second plurality of heterozygous loci for the first chromosomal region in the paternal genome, each of the second plurality of heterozygous loci including the corresponding third allele in the first paternal haplotype and the corresponding fourth allele in the second paternal haplotype, wherein the paternal genome corresponds to the father of the fetus; A third set of multiple free DNA molecules were identified, each of which: at any of the second plurality of heterozygous loci, includes the corresponding third allele of the heterozygous locus, and including at least one of the K genomic loci; Using the second plurality of free DNA molecular weights of the third set to measure the methylation degree of the K third mixture of the K genomic loci; determining the third percent contribution of the fetal tissue type in the mixture using the K third methylation levels; A fourth set of multiple free DNA molecules were identified, each of which: at any of the second plurality of heterozygous loci, includes the corresponding fourth allele of the heterozygous locus, and including at least one of the K genomic loci; measuring the degree of methylation of the K fourth mixtures of the K genomic loci using the second plurality of free DNA molecular weights of the fourth group; determining the fourth percent contribution of the fetal tissue type in the mixture using the K fourth methylation levels; calculating a second separation value between the third percent contribution and the fourth percent contribution; The portion of the fetal genome of the second plurality of heterozygous loci is determined based on the second segregation value. 如請求項2之方法，其中該第一分離值為該第一百分比貢獻與該第二百分比貢獻的比率，且其中當該比率等於閾值內之一時，該胎兒基因組之部分測定為具有該第一單倍型及該第二單倍型。The method of claim 2, wherein the first separation value is the ratio of the first percent contribution to the second percent contribution, and wherein when the ratio is equal to one within a threshold, the portion of the fetal genome is determined to be Has the first haplotype and the second haplotype. 如請求項2之方法，其中該第一分離值為該第一百分比貢獻與該第二百分比貢獻之差值。The method of claim 2, wherein the first separation value is a difference between the first percentage contribution and the second percentage contribution. 如請求項2之方法，其中當該第一分離值大於臨限值時，該胎兒基因組之部分測定為具有該第一單倍型之一或多個複本且無該第二單倍型之複本。The method of claim 2, wherein when the first segregation value is greater than a threshold value, the portion of the fetal genome is determined to have one or more copies of the first haplotype and no copies of the second haplotype . 如請求項2之方法，其中當該第一分離值小於臨限值時，該胎兒基因組之部分測定為具有該第二單倍型之一或多個複本且無該第一單倍型之複本。The method of claim 2, wherein when the first segregation value is less than a threshold value, the portion of the fetal genome is determined to have one or more copies of the second haplotype and no copies of the first haplotype . 如請求項1之方法，其中N為10或大於10。The method of claim 1, wherein N is 10 or greater. 如請求項1之方法，其中該N個混合物甲基化程度形成甲基化向量b，且其中測定該胎兒組織類型之該第一百分比貢獻包括： 對於M種組織類型中之每一者： 獲得該N個基因組位點之N個組織特異性甲基化程度，N大於或等於M，其中該等組織特異性甲基化程度形成維度N×M之矩陣A，該M種組織類型包括該胎兒組織類型； 對提供該矩陣A之甲基化向量b之組合向量x求解；及 對於該組合向量x之一或多個分量中之每一者： 使用該分量測定該混合物中該M種組織類型之對應組織類型之對應百分比貢獻。The method of claim 1, wherein the N mixture degrees of methylation form a methylation vector b, and wherein determining the first percent contribution of the fetal tissue type comprises: For each of the M tissue types: Obtain N tissue-specific methylation levels of the N genomic loci, where N is greater than or equal to M, wherein the tissue-specific methylation levels form a matrix A of dimension N×M, and the M tissue types include the fetal tissue type; solve for the combined vector x that provides the methylation vectors b of the matrix A; and For each of the one or more components of the combined vector x: The components are used to determine the corresponding percent contribution of the corresponding tissue types of the M tissue types in the mixture. 如請求項10之方法，其中該M種組織類型包括肝臟、肺、嗜中性白血球、淋巴球、紅血球母細胞、心臟、大腸、小腸及胎盤，其中該胎兒組織類型為胎盤。The method of claim 10, wherein the M tissue types include liver, lung, neutrophils, lymphocytes, erythroblasts, heart, large intestine, small intestine, and placenta, wherein the fetal tissue type is placenta. 如請求項1之方法，其中量測該N個基因組位點之N個第一混合物甲基化程度包括分析能識別甲基化感測(methylation-aware)定序結果，且其中該複數個游離之DNA分子之位置係使用該等能識別甲基化感測定序測序結果測定。The method of claim 1, wherein measuring the methylation levels of the N first mixtures of the N genomic loci comprises analyzing a methylation-aware sequencing result, and wherein the plurality of free The positions of the DNA molecules were determined using the sequencing results of these methylation-aware sensing sequences. 如請求項1之方法，其中該第一分離值包括該第一百分比貢獻與該第二百分比貢獻之比率。The method of claim 1, wherein the first separation value comprises a ratio of the first percent contribution to the second percent contribution. 如請求項1之方法，其中該第一分離值包括該第一百分比貢獻與該第二百分比貢獻之差值。The method of claim 1, wherein the first separation value comprises the difference between the first percent contribution and the second percent contribution. 一種使用來自懷孕女性之生物樣品測定該懷孕女性之未出生胎兒之胎兒基因組之一部分的方法，其中該生物樣品包括來自複數個組織類型，包括母體組織類型及胎兒組織類型之游離之DNA分子之混合物，該胎兒具有父親及母親為該懷孕女性，該方法包含： 藉由電腦系統分析來自該生物樣品之複數個游離之DNA分子，該複數個游離之DNA分子為至少1,000個游離之DNA分子，其中分析游離之DNA分子包括： 鑑別該游離之DNA分子於參考人類基因組中之位置；及 測定該游離之DNA分子之各別等位基因； 測定該胎兒之第一親體的第一親體基因組之第一染色體區域之第一單倍型； 鑑別該第一親體基因組之該第一染色體區域之一或多個雜合基因座，各雜合基因座包括該第一親體基因組之該第一染色體區域之該第一單倍型中之對應第一等位基因及第二單倍型中之對應第二等位基因； 鑑別第一組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該雜合基因座之該對應第一等位基因，且 包括N個基因組位點中之至少一者，N為大於或等於2之整數； 使用該第一組複數個游離之DNA分子量測該N個基因組位點之N個第一混合物甲基化程度； 藉由該電腦系統使用該N個第一甲基化程度測定該混合物中之該胎兒組織類型之第一百分比貢獻；及 比較該第一百分比貢獻與參考值以測定該胎兒是否在該第一染色體區域遺傳該第一單倍型，其中當該第一百分比貢獻超過該參考值時，該胎兒在該第一染色體區域遺傳該第一單倍型。A method of determining a portion of the fetal genome of an unborn fetus of a pregnant woman using a biological sample from a pregnant woman, wherein the biological sample includes a mixture of cell-free DNA molecules from a plurality of tissue types, including maternal and fetal tissue types , the fetus has father and mother as the pregnant woman, and the method includes: Analyzing a plurality of free DNA molecules from the biological sample by a computer system, the plurality of free DNA molecules is at least 1,000 free DNA molecules, wherein analyzing the free DNA molecules includes: identifying the location of the episomal DNA molecule in the reference human genome; and determining the individual alleles of the free DNA molecule; determining the first haplotype of the first chromosomal region of the first parent genome of the first parent of the fetus; Identifying one or more heterozygous loci in the first chromosomal region of the first parental genome, each heterozygous locus comprising the corresponding first haplotype in the first haplotype in the first chromosomal region of the first parental genome The first allele and the corresponding second allele in the second haplotype; A first set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding first allele of the heterozygous locus, and including at least one of N genomic loci, where N is an integer greater than or equal to 2; measuring the methylation degree of the N first mixtures of the N genomic loci using the first set of free DNA molecular weights; Determining, by the computer system, the first percent contribution of the fetal tissue type in the mixture using the N first methylation levels; and Comparing the first percent contribution to a reference value to determine whether the fetus inherits the first haplotype in the first chromosomal region, wherein when the first percent contribution exceeds the reference value, the fetus is in the first A chromosomal region inherits the first haplotype. 如請求項15之方法，其中該參考值係如下測定： 測定該第一親體基因組之第一染色體區域之第二單倍型； 鑑別第二組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該雜合基因座之該對應第二等位基因，且 包括該N個基因組位點中之至少一者； 使用該第二組複數個游離之DNA分子量測該N個基因組位點之N個第二混合物甲基化程度； 使用該N個第二甲基化程度量測該混合物中之該胎兒組織類型之第二百分比貢獻； 以該第二百分比貢獻及臨限值之總和測定該參考值。The method of claim 15, wherein the reference value is determined as follows: determining the second haplotype of the first chromosomal region of the first parent genome; A second set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding second allele of the heterozygous locus, and including at least one of the N genomic loci; measuring the degree of methylation of the N second mixtures of the N genomic loci using the second set of free DNA molecular weights; using the N second methylation levels to measure the second percent contribution of the fetal tissue type in the mixture; The reference value is determined as the sum of the second percent contribution and the threshold value. 如請求項15之方法，其進一步包含： 測定該第一親體基因組之第一染色體區域之第二單倍型； 鑑別第二組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該雜合基因座之該對應第二等位基因，且 包括該N個基因組位點中之至少一者； 使用該第二組複數個游離之DNA分子量測該N個基因組位點之N個第二混合物甲基化程度； 使用該N個第二甲基化程度測定該混合物中之該胎兒組織類型之第二百分比貢獻； 比較該第二百分比貢獻與參考值以測定該胎兒是否在該第一染色體區域遺傳該第二單倍型，其中當該第二百分比貢獻超過該參考值時，該胎兒在該第一染色體區域遺傳該第二單倍型。The method of claim 15, further comprising: determining the second haplotype of the first chromosomal region of the first parent genome; A second set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding second allele of the heterozygous locus, and including at least one of the N genomic loci; measuring the degree of methylation of the N second mixtures of the N genomic loci using the second set of free DNA molecular weights; determining the second percent contribution of the fetal tissue type in the mixture using the N second methylation levels; Comparing the second percent contribution to a reference value to determine whether the fetus inherits the second haplotype in the first chromosomal region, wherein when the second percent contribution exceeds the reference value, the fetus is in the first chromosomal region. A chromosomal region inherits the second haplotype. 一種使用來自懷孕女性之生物樣品測定該懷孕女性之未出生胎兒之胎兒基因組之一部分的方法，其中該生物樣品包括來自複數個組織類型，包括母體組織類型及胎兒組織類型之游離之DNA分子之混合物，該胎兒具有父親及母親為該懷孕女性，該方法包含： 藉由電腦系統分析來自該生物樣品之複數個游離之DNA分子，該複數個游離之DNA分子為至少1,000個游離之DNA分子，其中分析游離之DNA分子包括： 鑑別該游離之DNA分子於參考人類基因組中之位置；及 測定該游離之DNA分子之各別等位基因； 測定該胎兒之第一親體的第一親體基因組之第一染色體區域之第一單倍型及第二單倍型； 鑑別該第一親體基因組之該第一染色體區域之複數個雜合基因座，各雜合基因座包括該第一單倍型中之對應第一等位基因及該第二單倍型中之對應第二等位基因； 鑑別第一組複數個游離之DNA分子，其各自： 位於該複數個雜合基因座中的任一者， 包括該雜合基因座之該對應第一等位基因，且 包括N個基因組位點中之至少一者； 使用該第一組複數個游離之DNA分子量測第一混合物甲基化程度； 鑑別第二組複數個游離之DNA分子，其各自： 位於該複數個雜合基因座中的任一者， 包括該雜合基因座之該對應第二等位基因，且 包括該N個基因組位點中之至少一者； 使用該第二組複數個游離之DNA分子量測第二混合物甲基化程度；及 基於該第一混合物甲基化程度及該第二混合物甲基化程度中之何者較低來測定該胎兒遺傳該第一單倍型及該第二單倍型中之何者。A method of determining a portion of the fetal genome of an unborn fetus of a pregnant woman using a biological sample from a pregnant woman, wherein the biological sample includes a mixture of cell-free DNA molecules from a plurality of tissue types, including maternal and fetal tissue types , the fetus has father and mother as the pregnant woman, and the method includes: Analyzing a plurality of free DNA molecules from the biological sample by a computer system, the plurality of free DNA molecules is at least 1,000 free DNA molecules, wherein analyzing the free DNA molecules includes: identifying the location of the episomal DNA molecule in the reference human genome; and determining the individual alleles of the free DNA molecule; determining the first haplotype and the second haplotype of the first chromosomal region of the first parent genome of the first parent of the fetus; identifying a plurality of heterozygous loci in the first chromosomal region of the first parental genome, each heterozygous locus comprising a corresponding first allele in the first haplotype and a corresponding one in the second haplotype second allele; A first set of multiple free DNA molecules are identified, each of which: at any of the plurality of heterozygous loci, includes the corresponding first allele of the heterozygous locus, and including at least one of the N genomic loci; using the first set of free DNA molecular weights to measure the methylation level of the first mixture; A second set of multiple free DNA molecules are identified, each of which: at any of the plurality of heterozygous loci, includes the corresponding second allele of the heterozygous locus, and including at least one of the N genomic loci; using the second set of free DNA molecular weights to measure the degree of methylation of the second mixture; and Which of the first haplotype and the second haplotype is inherited by the fetus is determined based on which of the degree of methylation of the first mixture and the degree of methylation of the second mixture is lower. 如請求項18之方法，其中該第一混合物甲基化程度為該第一組複數個游離之DNA分子之甲基化密度，且其中該第二混合物甲基化程度為該第二組複數個游離之DNA分子之甲基化密度。The method of claim 18, wherein the degree of methylation of the first mixture is the methylation density of the first plurality of free DNA molecules, and wherein the degree of methylation of the second mixture is the degree of methylation of the second plurality of Methylation density of free DNA molecules. 如請求項18之方法，其中測定遺傳該第一單倍型及該第二單倍型中之何者包括： 測定該第一混合物甲基化程度與該第二混合物甲基化程度之間的分離值；及 比較該分離值與臨限值。The method of claim 18, wherein determining which of the first haplotype and the second haplotype is inherited comprises: determining the separation value between the degree of methylation of the first mixture and the degree of methylation of the second mixture; and Compare this separation value with a threshold value. 如請求項20之方法，其中該臨限值係使用斯圖登氏t檢驗(Student's t-test)、曼-惠氏秩和檢驗(Mann-Whitney rank-sum test)或卡方檢驗(Chi-square test)測定。The method of claim 20, wherein the threshold value is determined using a Student's t-test, a Mann-Whitney rank-sum test, or a Chi-square test test) determination. 如請求項18之方法，其中測定該第一親體基因組之第一染色體區域之第一單倍型及第二單倍型包括： 使用來自該第一親體之樣品測定該複數個雜合基因座之第一親體基因組之基因型； 獲得複數個參考單倍型；及 使用該等基因型及該複數個參考單倍型推論該第一單倍型及該第二單倍型。The method of claim 18, wherein determining the first haplotype and the second haplotype of the first chromosomal region of the first parental genome comprises: Genotyping the genome of the first parent of the plurality of heterozygous loci using the sample from the first parent; obtaining a plurality of reference haplotypes; and The first haplotype and the second haplotype are inferred using the genotypes and the plurality of reference haplotypes. 一種使用來自懷孕女性之生物樣品偵測該懷孕女性之未出生胎兒之胎兒基因組之一部分中之序列不平衡的方法，其中該生物樣品包括來自複數個組織類型，包括母體組織類型及胎兒組織類型之游離之DNA分子之混合物，該胎兒具有父親及母親為該懷孕女性，該方法包含： 藉由電腦系統分析來自該生物樣品之複數個游離之DNA分子，該複數個游離之DNA分子為至少1,000個游離之DNA分子，其中分析游離之DNA分子包括： 鑑別該游離之DNA分子於參考人類基因組中之位置；及 測定該游離之DNA分子之各別等位基因； 測定該胎兒之第一親體的第一親體基因組之目標染色體區域之第一目標單倍型； 鑑別該第一親體基因組之目標染色體區域之複數個目標雜合基因座，各目標雜合基因座包括該第一親體基因組之目標染色體區域之第一目標單倍型中之對應第一目標等位基因及第二目標單倍型中之對應第二目標等位基因； 鑑別目標組之複數個游離之DNA分子，其各自： 位於該目標雜合基因座中的任一者， 包括該目標雜合基因座之對應第一目標等位基因，且 包括該目標染色體區域中之N個基因組位點中之至少一者，N為大於或等於2之整數； 使用該目標組之複數個游離之DNA分子量測該N個基因組位點之N個目標混合物甲基化程度； 藉由該電腦系統使用該N個目標甲基化程度測定該混合物中之該胎兒組織類型之第一目標百分比貢獻；及 測定該第一親體基因組之參考染色體區域之第一參考單倍型，該參考染色體區域不同於該目標染色體區域； 鑑別該第一親體基因組之參考染色體區域之複數個參考雜合基因座，各參考雜合基因座包括該第一親體基因組之參考染色體區域之該第一參考單倍型中之對應第一參考等位基因及第二參考單倍型中之對應第二參考等位基因； 鑑別參考組之複數個游離之DNA分子，其各自： 位於該參考雜合基因座中的任一者， 包括該參考雜合基因座之該對應第一參考等位基因，且 包括該參考染色體區域中之K個基因組位點中之至少一者； 使用該參考組之複數個游離之DNA分子量測該K個基因組位點之K個參考混合物甲基化程度； 藉由該電腦系統使用該K個參考甲基化程度測定該混合物中之該胎兒組織類型之第一參考百分比貢獻； 計算該第一目標百分比貢獻與該第一參考百分比貢獻之間的第一分離值；及 比較該第一分離值與臨限值以確定該胎兒是否對於該目標染色體區域具有該序列不平衡之分類。A method of using a biological sample from a pregnant woman to detect sequence imbalances in a portion of the fetal genome of the pregnant woman's unborn fetus, wherein the biological sample comprises from a plurality of tissue types, including the difference between a maternal tissue type and a fetal tissue type The mixture of free DNA molecules, the fetus has the father and the mother is the pregnant female, the method comprises: Analyzing a plurality of free DNA molecules from the biological sample by a computer system, the plurality of free DNA molecules is at least 1,000 free DNA molecules, wherein analyzing the free DNA molecules includes: identifying the location of the episomal DNA molecule in the reference human genome; and determining the individual alleles of the free DNA molecule; determining the first target haplotype of the target chromosomal region of the first parent genome of the first parent of the fetus; Identifying a plurality of target heterozygous loci in the target chromosomal region of the first parental genome, each target heterozygous locus includes a corresponding first target allele in the first target haplotype in the target chromosomal region of the first parental genome the corresponding second target allele in the gene and the second target haplotype; A plurality of free DNA molecules of the target group are identified, each of which: at any of the target heterozygous loci, includes the corresponding first target allele for the target heterozygous locus, and Include at least one of N genomic loci in the target chromosomal region, where N is an integer greater than or equal to 2; Using the free DNA molecular weights of the target group to measure the methylation degree of the N target mixtures of the N genomic loci; determining, by the computer system, the first target percent contribution of the fetal tissue type in the mixture using the N target methylation levels; and determining a first reference haplotype of a reference chromosomal region of the first parental genome, the reference chromosomal region being different from the target chromosomal region; Identifying a plurality of reference heterozygous loci of the reference chromosomal region of the first parental genome, each reference heterozygous locus comprising a corresponding first reference in the first reference haplotype of the reference chromosomal region of the first parental genome, etc. the corresponding second reference allele in the allele and the second reference haplotype; A plurality of free DNA molecules of the reference set are identified, each of which: at any of the reference heterozygous loci, includes the corresponding first reference allele of the reference heterozygous locus, and comprising at least one of the K genomic loci in the reference chromosomal region; Using the free DNA molecular weights of the reference set to measure the methylation levels of the K reference mixtures of the K genomic loci; Determining, by the computer system, the first reference percent contribution of the fetal tissue type in the mixture using the K reference methylation levels; calculating a first separation value between the first target percentage contribution and the first reference percentage contribution; and The first segregation value is compared to a threshold value to determine whether the fetus has the classification of the sequence imbalance for the target chromosomal region. 如請求項23之方法，其中該序列不平衡為該目標染色體區域之擴增。The method of claim 23, wherein the sequence imbalance is an amplification of the target chromosomal region. 如請求項23之方法，其中使用之臨限值取決於測試之序列不平衡之類型。The method of claim 23, wherein the threshold value used depends on the type of sequence imbalance being tested. 如請求項23之方法，其進一步包含： 量測該生物樣品中之胎兒DNA百分比；及 使用該胎兒DNA百分比測定該臨限值。The method of claim 23, further comprising: measuring the percentage of fetal DNA in the biological sample; and The threshold value was determined using the fetal DNA percentage. 如請求項23之方法，其進一步包含： 使用參考組之樣品測定該臨限值，該等樣品均具有該序列不平衡或不具有該序列不平衡。The method of claim 23, further comprising: The threshold value was determined using samples of the reference set, all of which either had the sequence imbalance or did not have the sequence imbalance. 如請求項23之方法，其進一步包含： 測定該胎兒自該第一親體遺傳該第一目標單倍型；及 測定該胎兒自該第一親體遺傳該第一參考單倍型。The method of claim 23, further comprising: determining that the fetus has inherited the first target haplotype from the first parent; and It is determined that the fetus has inherited the first reference haplotype from the first parent. 如請求項28之方法，其中測定該胎兒自該第一親體遺傳該第一目標單倍型包括： 測定對應於該第二目標單倍型之混合物中該胎兒組織類型之第二目標百分比貢獻； 計算該第一目標百分比貢獻與該第二目標百分比貢獻之間的第二分離值； 基於該第二分離值測定該胎兒自該第一親體遺傳該第一目標單倍型。The method of claim 28, wherein determining that the fetus has inherited the first target haplotype from the first parent comprises: determining the second target percent contribution of the fetal tissue type in the mixture corresponding to the second target haplotype; calculating a second separation value between the first target percentage contribution and the second target percentage contribution; It is determined that the fetus has inherited the first target haplotype from the first parent based on the second segregation value. 如請求項23之方法，其中該目標染色體區域為第一染色體且該參考染色體區域為不同於該第一染色體之第二染色體。The method of claim 23, wherein the target chromosomal region is a first chromosome and the reference chromosomal region is a second chromosome different from the first chromosome. 如請求項23之方法，其中該第一親體為該胎兒之母親。The method of claim 23, wherein the first parent is the mother of the fetus. 一種分析生物體之生物樣品之方法，該生物樣品包括來自複數種組織類型，包括第一組織類型之游離之DNA分子的混合物，該方法包含： 藉由電腦系統分析來自該生物樣品之複數個游離之DNA分子，該複數個游離之DNA分子為至少1,000個游離之DNA分子，其中分析游離之DNA分子包括： 鑑別對應於該生物體之參考基因組中之該游離之DNA分子之位置； 測定該游離之DNA分子之各別等位基因； 鑑別一或多個基因座，其各自具有第一染色體區域之第一單倍型上之對應第一等位基因，其中該第一單倍型： 不存在該生物體之健康細胞中，或 存在該生物體之健康細胞中且不存在可能在該混合物中之異常細胞中； 鑑別第一組複數個游離之DNA分子，其各自： 位於該一或多個基因座之任一基因座， 包括該基因座之對應第一等位基因，且 包括N個基因組位點中之至少一者，N為大於或等於2之整數； 使用該第一組複數個游離之DNA分子量測該N個基因組位點之N個第一混合物甲基化程度； 藉由該電腦系統使用該N個第一甲基化程度測定該混合物中之該第一組織類型之第一百分比貢獻； 測定該第一百分比貢獻與參考百分比貢獻之間的第一分離值；及 比較該第一分離值與臨限值以確定該第一組織類型是否具有疾病病況之分類。A method of analyzing a biological sample of an organism, the biological sample comprising a mixture of free DNA molecules from a plurality of tissue types, including a first tissue type, the method comprising: Analyzing a plurality of free DNA molecules from the biological sample by a computer system, the plurality of free DNA molecules is at least 1,000 free DNA molecules, wherein analyzing the free DNA molecules includes: identifying the location corresponding to the episomal DNA molecule in the organism's reference genome; determining the individual alleles of the free DNA molecule; Identifying one or more loci, each having a corresponding first allele on a first haplotype of a first chromosomal region, wherein the first haplotype: not present in healthy cells of the organism, or present in healthy cells of the organism and not in abnormal cells that may be in the mixture; A first set of multiple free DNA molecules are identified, each of which: at any of the one or more loci, includes the corresponding first allele of the locus, and including at least one of N genomic loci, where N is an integer greater than or equal to 2; measuring the methylation degree of the N first mixtures of the N genomic loci using the first set of free DNA molecular weights; determining, by the computer system, the first percent contribution of the first tissue type in the mixture using the N first methylation levels; determining a first separation value between the first percent contribution and the reference percent contribution; and The first separation value is compared to a threshold value to determine whether the first tissue type has a classification of a disease condition. 如請求項32之方法，其中該第一單倍型在第一染色體上，且其中該第一組織類型在該第一染色體之兩個複本上均具有該第一單倍型。The method of claim 32, wherein the first haplotype is on a first chromosome, and wherein the first tissue type has the first haplotype on both copies of the first chromosome. 如請求項32之方法，其中該第一單倍型不存在該生物體之健康細胞中，且其中該第一單倍型與該疾病病況相關。The method of claim 32, wherein the first haplotype is not present in healthy cells of the organism, and wherein the first haplotype is associated with the disease condition. 如請求項34之方法，其中當該第一分離值大於該臨限值時，該第一組織類型測定為具有該疾病病況。The method of claim 34, wherein the first tissue type is determined to have the disease condition when the first separation value is greater than the threshold value. 如請求項34之方法，其進一步包含： 鑑別第二組複數個游離之DNA分子，其各自： 位於該一或多個基因座中的任一者； 包括該第一染色體區域之第二單倍型上之對應第二等位基因，該第二單倍型不存在該生物體之健康細胞中，且該第二單倍型與該疾病病況相關；且 包括該N個基因組位點中之至少一者； 使用該第二組複數個游離之DNA分子量測該N個基因組位點之N個第二混合物甲基化程度； 藉由該電腦系統使用該N個第二甲基化程度測定該混合物中之該第一組織類型之第二百分比貢獻； 測定該第二百分比貢獻與該參考百分比貢獻之間的第二分離值；及 比較該第二分離值與該臨限值作為確定該第一組織類型是否具有該疾病病況之分類之一部分。The method of claim 34, further comprising: A second set of multiple free DNA molecules are identified, each of which: at any of the one or more loci; comprising a corresponding second allele on a second haplotype of the first chromosomal region that is not present in healthy cells of the organism and that is associated with the disease condition; and including at least one of the N genomic loci; measuring the degree of methylation of the N second mixtures of the N genomic loci using the second set of free DNA molecular weights; determining, by the computer system, the second percent contribution of the first tissue type in the mixture using the N second methylation levels; determining a second separation value between the second percent contribution and the reference percent contribution; and The second separation value is compared to the threshold value as part of the classification of determining whether the first tissue type has the disease condition. 如請求項34之方法，其中該疾病病況為癌症。The method of claim 34, wherein the disease condition is cancer. 如請求項37之方法，其中特定癌症在該特定癌症之癌症基因組中具有該第一單倍型。The method of claim 37, wherein a specific cancer has the first haplotype in the cancer genome of the specific cancer. 如請求項38之方法，其中該特定癌症選自由以下組成之群：肝癌、肺癌、胰臟癌、心房癌、結腸癌、乙狀結腸癌、橫結腸癌、升結腸癌、降結腸癌、腎上腺癌、食道癌、小腸癌及CD4 T細胞癌。The method of claim 38, wherein the specific cancer is selected from the group consisting of liver cancer, lung cancer, pancreatic cancer, atrial cancer, colon cancer, sigmoid colon cancer, transverse colon cancer, ascending colon cancer, descending colon cancer, adrenal cancer, esophagus cancer cancer, small bowel cancer, and CD4 T-cell cancer. 如請求項37之方法，其中該第一組織類型是否具有癌症之分類包含存在或不存在癌症之分類、癌症階段之分類、腫瘤尺寸之分類及/或轉移之分類。The method of claim 37, wherein the classification of whether the first tissue type has cancer includes classification of the presence or absence of cancer, classification of cancer stage, classification of tumor size, and/or classification of metastasis. 如請求項37之方法，其進一步包含藉由該電腦系統執行以下： 使用該等N個第一混合物甲基化程度測定該混合物中之其他組織類型之複數個對應百分比貢獻； 測定該複數個對應百分比貢獻與對應參考百分比貢獻之間的對應分離值；及 比較該對應分離值與該臨限值以確定該等其他組織類型中之每一者是否具有該特定癌症之分類。The method of claim 37, further comprising performing, by the computer system, the following: using the N first mixture methylation levels to determine a plurality of corresponding percentage contributions of other tissue types in the mixture; determining the corresponding separation value between the plurality of corresponding percentage contributions and the corresponding reference percentage contributions; and The corresponding segregation value is compared to the threshold value to determine whether each of the other tissue types has the classification of the particular cancer. 如請求項34之方法，其中該一或多個基因座中之每一者之該對應第一等位基因為癌症特異性突變。The method of claim 34, wherein the corresponding first allele of each of the one or more loci is a cancer-specific mutation. 如請求項32之方法，其進一步包含： 如下測定該第一單倍型上之該一或多個基因座中之每一者之對應第一等位基因： 分析具有該第一單倍型之組織樣品之DNA分子以測定該第一單倍型。The method of claim 32, further comprising: The corresponding first allele for each of the one or more loci on the first haplotype is determined as follows: DNA molecules of the tissue sample having the first haplotype are analyzed to determine the first haplotype. 如請求項32之方法，其中該疾病病況為該第一組織類型之移植體於該生物體中排斥。The method of claim 32, wherein the disease condition is rejection of a transplant of the first tissue type in the organism. 如請求項44之方法，其中該參考百分比貢獻自該第一組織類型為健康之生物體之生物樣品之一或多次量測測定。The method of claim 44, wherein the reference percentage is contributed from one or more measurements of a biological sample of an organism in which the first tissue type is healthy. 如請求項44之方法，其中該參考百分比貢獻自移植之第一組織不遭排斥之生物體之生物樣品之一或多次量測測定。The method of claim 44, wherein the reference percentage contributes to one or more measurements from a biological sample of an organism whose transplanted first tissue is not rejected. 如請求項32之方法，其中該參考百分比貢獻為零。The method of claim 32, wherein the reference percentage contribution is zero. 如請求項32之方法，其中該等N個第一混合物甲基化程度形成甲基化向量b，且其中測定該第一組織類型之第一百分比貢獻包括： 對於M種組織類型中之每一者： 獲得該N個基因組位點之N個組織特異性甲基化程度，N大於或等於M，其中該等組織特異性甲基化程度形成維度N×M之矩陣A，該M種組織類型包括該第一組織類型，M為大於1之整數； 對提供該矩陣A之甲基化向量b之組合向量x求解 對於該組合向量x之一或多個分量中之各分量： 使用該分量測定該混合物中該M種組織類型之對應組織類型之對應百分比貢獻。The method of claim 32, wherein the N degrees of methylation of the first mixture form a methylation vector b, and wherein determining the first percent contribution of the first tissue type comprises: For each of the M tissue types: Obtain N tissue-specific methylation levels of the N genomic loci, where N is greater than or equal to M, wherein the tissue-specific methylation levels form a matrix A of dimension N×M, and the M tissue types include the The first tissue type, M is an integer greater than 1; Solve for the combined vector x that provides the methylation vector b of this matrix A For each of the one or more components of the combined vector x: The components are used to determine the corresponding percent contribution of the corresponding tissue types of the M tissue types in the mixture. 如請求項32之方法，其中該第一單倍型存在該生物體之健康細胞中且不存在可能在該混合物中之異常細胞中。The method of claim 32, wherein the first haplotype is present in healthy cells of the organism and absent in abnormal cells that may be in the mixture. 如請求項32之方法，其中當該第一分離值小於該臨限值時，該第一組織類型測定為具有該疾病病況。The method of claim 32, wherein the first tissue type is determined to have the disease condition when the first separation value is less than the threshold value. 如請求項49之方法，其中該等異常細胞來自腫瘤。The method of claim 49, wherein the abnormal cells are from a tumor. 如請求項49之方法，其中該等異常細胞來自供體組織。The method of claim 49, wherein the abnormal cells are from donor tissue. 一種分析生物體之生物樣品以鑑別染色體畸變(aberration)之起源的方法，該生物樣品包括來自複數種組織類型，包括第一組織類型之游離之DNA分子之混合物，該方法包含： 藉由電腦系統分析來自該生物樣品之複數個游離之DNA分子，該複數個游離之DNA分子為至少1,000個游離之DNA分子，其中分析游離之DNA分子包括： 鑑別對應於該生物體之參考基因組中該游離之DNA分子之位置； 測定該游離之DNA分子之各別等位基因； 基於位於第一染色體區域中之游離之DNA分子之第一量將該第一染色體區域鑑別為在該生物體中展現拷貝數變異； 測定該第一染色體區域中之該生物體之第一單倍型及第二單倍型； 鑑別該第一染色體區域之一或多個雜合基因座，各雜合基因座包括該第一單倍型中之對應第一等位基因及該第二單倍型中之對應第二等位基因； 鑑別第一組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該雜合基因座之該對應第一等位基因，且 包括N個基因組位點中之至少一者，N為大於或等於2之整數； 使用該第一組複數個游離之DNA分子量測該N個基因組位點之N個第一混合物甲基化程度； 鑑別第二組複數個游離之DNA分子，其各自： 位於該一或多個雜合基因座中的任一者， 包括該雜合基因座之該對應第二等位基因，且 包括該N個基因組位點中之至少一者； 使用該第二組複數個游離之DNA分子量測該N個基因組位點之N個第二混合物甲基化程度； 對於M種組織類型中之各組織類型： 藉由該電腦系統使用該N個第一甲基化程度測定該混合物中該組織類型之對應第一百分比貢獻，M為大於1之整數； 藉由該電腦系統使用該N個第二甲基化程度測定該混合物中該組織類型之對應第二百分比貢獻； 計算該對應第一百分比貢獻與該對應第二百分比貢獻之間的對應分離值； 基於該等對應分離值中具有最大值之第一組織類型之第一分離值將該第一組織類型鑑別為拷貝數變異之起源。A method of analyzing a biological sample of an organism to identify the origin of chromosomal aberrations, the biological sample comprising a mixture of free DNA molecules from a plurality of tissue types, including a first tissue type, the method comprising: Analyzing a plurality of free DNA molecules from the biological sample by a computer system, the plurality of free DNA molecules is at least 1,000 free DNA molecules, wherein analyzing the free DNA molecules includes: identifying the location corresponding to the episomal DNA molecule in the organism's reference genome; determining the individual alleles of the free DNA molecule; Identifying the first chromosomal region as exhibiting copy number variation in the organism based on a first amount of free DNA molecules located in the first chromosomal region; determining a first haplotype and a second haplotype of the organism in the first chromosomal region; Identifying one or more heterozygous loci in the first chromosomal region, each heterozygous locus comprising a corresponding first allele in the first haplotype and a corresponding second allele in the second haplotype Gene; A first set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding first allele of the heterozygous locus, and including at least one of N genomic loci, where N is an integer greater than or equal to 2; measuring the methylation degree of the N first mixtures of the N genomic loci using the first set of free DNA molecular weights; A second set of multiple free DNA molecules are identified, each of which: at any of the one or more heterozygous loci, includes the corresponding second allele of the heterozygous locus, and including at least one of the N genomic loci; measuring the degree of methylation of the N second mixtures of the N genomic loci using the second set of free DNA molecular weights; For each of the M tissue types: Determining, by the computer system, the corresponding first percent contribution of the tissue type in the mixture using the N first methylation levels, where M is an integer greater than 1; Determining, by the computer system, the corresponding second percent contribution of the tissue type in the mixture using the N second degrees of methylation; calculating a corresponding separation value between the corresponding first percentage contribution and the corresponding second percentage contribution; The first tissue type is identified as the origin of the copy number variation based on the first segregation value of the first tissue type having the largest value among the corresponding segregation values. 如請求項53之方法，其進一步包含： 測定該第一分離值與次最高(next highest)分離值之間的差值；及 比較該差值與臨限值以確定該第一組織類型為該拷貝數變異之起源之可能性的分類。The method of claim 53, further comprising: determining the difference between the first separation value and the next highest separation value; and The difference is compared to a threshold value to determine the classification of the likelihood that the first tissue type is the origin of the copy number variation. 如請求項53之方法，其中將該第一染色體區域鑑別為展現該拷貝數變異包括： 比較位於該第一染色體區域中之游離之DNA分子之第一量與位於其他染色體區域中之游離之DNA分子之量測定的參考值。The method of claim 53, wherein identifying the first chromosomal region as exhibiting the copy number variation comprises: A reference value determined by comparing the first amount of free DNA molecules located in the first chromosomal region with the amount of free DNA molecules located in other chromosomal regions. 如請求項53之方法，其中該拷貝數變異為複本數增加。The method of claim 53, wherein the copy number variation is an increase in the number of copies. 如請求項53之方法，其進一步包含： 測定該第一組複數個游離之DNA分子中之游離之DNA分子之第一數目； 測定該第二組複數個游離之DNA分子中之游離之DNA分子之第二數目；及 使用該第一數目及該第二數目測定該最大值為正或負。The method of claim 53, further comprising: determining a first number of free DNA molecules in the first set of the plurality of free DNA molecules; determining a second number of free DNA molecules in the second set of free DNA molecules; and The maximum value is determined to be positive or negative using the first number and the second number. 如請求項53之方法，其中分析該游離之DNA分子包括：測定游離之DNA分子之尺寸，其中該第一組複數個游離之DNA分子具有第一尺寸分佈，且其中該第二組複數個游離之DNA分子具有第二尺寸分佈，該方法進一步包含： 測定該第一尺寸分佈之第一統計值； 測定該第二尺寸分佈之第二統計值；及 使用該第一統計值及該第二統計值測定該最大值為正或負。The method of claim 53, wherein analyzing the free DNA molecules comprises: determining the size of the free DNA molecules, wherein the first plurality of free DNA molecules have a first size distribution, and wherein the second plurality of free DNA molecules have a first size distribution the DNA molecules have a second size distribution, the method further comprising: determining a first statistic of the first size distribution; determining a second statistic of the second size distribution; and Whether the maximum value is positive or negative is determined using the first statistic and the second statistic. 如請求項58之方法，其進一步包含： 使用該第一統計值及該第二統計值測定該第一單倍型及該第二單倍型中之何者具有較高複本數；及 計算該等對應分離值，使得該第一分離值為正。The method of claim 58, further comprising: using the first statistic and the second statistic to determine which of the first haplotype and the second haplotype has the higher number of copies; and The corresponding separation values are calculated such that the first separation value is positive. 如請求項1至59中任一項之方法，其中分析該游離之DNA分子包含亞硫酸氫鹽定序、定序前甲基化敏感限制酶消化、使用抗甲基胞嘧啶抗體或甲基化結合蛋白之免疫沈澱，或可說明甲基化狀態之單分子定序。The method of any one of claims 1 to 59, wherein analyzing the free DNA molecule comprises bisulfite sequencing, pre-sequencing methylation-sensitive restriction enzyme digestion, use of anti-methylcytosine antibodies, or methylation Immunoprecipitation of binding proteins, or single-molecule sequencing that may account for methylation status. 一種電腦產品，其包含電腦可讀媒體，儲存複數個指令，用於控制電腦系統以執行如請求項1至59之任何方法之操作。A computer product comprising a computer-readable medium storing a plurality of instructions for controlling a computer system to perform operations as in any of the methods of claims 1-59. 一種系統，其包含： 如請求項61之電腦產品；及 一或多個處理器，其用於執行儲存於該電腦可讀媒體上的指令。A system comprising: Computer products as claimed in Item 61; and One or more processors for executing instructions stored on the computer-readable medium. 一種系統，其包含用於執行如請求項1至59之任何方法之構件。A system comprising means for performing any of the methods of claims 1-59. 一種系統，其經組態以執行如請求項1至59之任何方法。A system configured to perform any of the methods of claims 1-59. 一種系統，其包含分別執行如請求項1至59之任何方法之步驟的模組。A system comprising modules for performing the steps of any of the methods of claims 1-59, respectively.
TW110121521A 2015-07-20 2016-07-20 Methylation pattern analysis of haplotypes in tissues in a dna mixture TWI798718B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201562194702P 2015-07-20 2015-07-20
US62/194,702 2015-07-20

Publications (2)

Publication Number Publication Date
TW202205300A true TW202205300A (en) 2022-02-01
TWI798718B TWI798718B (en) 2023-04-11

Family

ID=57833669

Family Applications (2)

Application Number Title Priority Date Filing Date
TW110121521A TWI798718B (en) 2015-07-20 2016-07-20 Methylation pattern analysis of haplotypes in tissues in a dna mixture
TW105122965A TWI732771B (en) 2015-07-20 2016-07-20 Methylation pattern analysis of haplotypes in tissues in a dna mixture

Family Applications After (1)

Application Number Title Priority Date Filing Date
TW105122965A TWI732771B (en) 2015-07-20 2016-07-20 Methylation pattern analysis of haplotypes in tissues in a dna mixture

Country Status (12)

Country Link
US (2) US10689706B2 (en)
EP (3) EP3325663B1 (en)
CN (2) CN115287348A (en)
AU (2) AU2016295712B2 (en)
CA (1) CA2993588A1 (en)
DK (1) DK3739061T3 (en)
ES (1) ES2911613T3 (en)
HK (2) HK1248283A1 (en)
HU (1) HUE059407T2 (en)
PT (1) PT3739061T (en)
TW (2) TWI798718B (en)
WO (1) WO2017012544A1 (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106795562B (en) 2014-07-18 2022-03-25 香港中文大学 Tissue methylation pattern analysis in DNA mixtures
KR20230062684A (en) 2016-11-30 2023-05-09 더 차이니즈 유니버시티 오브 홍콩 Analysis of cell-free dna in urine and other samples
MY197535A (en) 2017-01-25 2023-06-21 Univ Hong Kong Chinese Diagnostic applications using nucleic acid fragments
WO2018236911A1 (en) * 2017-06-20 2018-12-27 Illumina, Inc. Methods and systems for decomposition and quantification of dna mixtures from multiple contributors of known or unknown genotypes
JP2020527958A (en) * 2017-07-26 2020-09-17 ザ チャイニーズ ユニバーシティー オブ ホンコンＴｈｅ Ｃｈｉｎｅｓｅ Ｕｎｉｖｅｒｓｉｔｙ Ｏｆ Ｈｏｎｇｋｏｎｇ Strengthening cancer screening using cell-free viral nucleic acids
WO2019159184A1 (en) * 2018-02-18 2019-08-22 Yissum Research Development Company Of The Hebrew University Of Jerusalem Ltd. Cell free dna deconvolution and use thereof
CA3105349A1 (en) 2018-05-03 2019-11-07 The Chinese University Of Hong Kong Size-tagged preferred ends and orientation-aware analysis for measuring properties of cell-free mixtures
BR112020023296A2 (en) * 2018-05-16 2021-05-04 Twinstrand Biosciences, Inc. methods and reagents for resolving mixtures of nucleic acid and mixed cell populations and associated applications
CA3111887A1 (en) 2018-09-27 2020-04-02 Grail, Inc. Methylation markers and targeted methylation probe panel
CN113661542A (en) * 2018-12-18 2021-11-16 格里尔公司 System and method for estimating cell-derived fraction using methylation information
EP3938534A4 (en) * 2019-03-13 2023-03-29 Grail, LLC Systems and methods for enriching for cancer-derived fragments using fragment size
CN116875669A (en) 2019-08-16 2023-10-13 香港中文大学 Determination of base modification of nucleic acids
WO2021077063A1 (en) * 2019-10-18 2021-04-22 Washington University Methods and systems for measuring cell states
KR20230113840A (en) * 2020-02-05 2023-08-01 더 차이니즈 유니버시티 오브 홍콩 Molecular analyses using long cell-free fragments in pregnancy
CN114566220A (en) * 2020-11-27 2022-05-31 深圳华大生命科学研究院 System for determining sample type based on DNA methylation level, readable medium and application thereof

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2514842B1 (en) 2007-07-23 2016-02-24 The Chinese University of Hong Kong Diagnosing fetal chromosomal aneuploidy using genomic sequencing
US20100112590A1 (en) 2007-07-23 2010-05-06 The Chinese University Of Hong Kong Diagnosing Fetal Chromosomal Aneuploidy Using Genomic Sequencing With Enrichment
US8563242B2 (en) * 2009-08-11 2013-10-22 The Chinese University Of Hong Kong Method for detecting chromosomal aneuploidy
SI3783110T1 (en) * 2009-11-05 2023-02-28 The Chinese University Of Hong Kong, Fetal genomic analysis from a maternal biological sample
EP3406737B1 (en) 2009-11-06 2023-05-31 The Chinese University of Hong Kong Size-based genomic analysis
TWI719339B (en) 2010-11-30 2021-02-21 香港中文大學 Detection of genetic or molecular aberrations associated with cancer
JP6105485B2 (en) * 2011-01-05 2017-04-05 ザ・チャイニーズ・ユニバーシティー・オブ・ホンコンＴｈｅ Ｃｈｉｎｅｓｅ Ｕｎｉｖｅｒｓｉｔｙ Ｏｆ Ｈｏｎｇ Ｋｏｎｇ Noninvasive prenatal identification of fetal sex chromosome genotypes
CA2827873C (en) * 2011-02-24 2022-08-16 The Chinese University Of Hong Kong Molecular testing of multiple pregnancies
US11261494B2 (en) 2012-06-21 2022-03-01 The Chinese University Of Hong Kong Method of measuring a fractional concentration of tumor DNA
SG11201501927VA (en) * 2012-09-20 2015-04-29 Univ Hong Kong Chinese Non-invasive determination of methylome of fetus or tumor from plasma
US9732390B2 (en) 2012-09-20 2017-08-15 The Chinese University Of Hong Kong Non-invasive determination of methylome of fetus or tumor from plasma
US10106836B2 (en) 2013-03-15 2018-10-23 The Chinese University Of Hong Kong Determining fetal genomes for multiple fetus pregnancies
US20150004601A1 (en) 2013-06-28 2015-01-01 Ariosa Diagnostics, Inc. Massively parallel sequencing of random dna fragments for determination of fetal fraction
EP3099822A4 (en) 2014-01-30 2017-08-30 The Regents of the University of California Methylation haplotyping for non-invasive diagnosis (monod)
PT4026917T (en) 2014-04-14 2024-02-12 Yissum Research And Development Company Of The Hebrew Univ Of Jerusalem Ltd A method and kit for determining the death of cells or tissue or the tissue or cell origin of dna by dna methylation analysis
CN106795562B (en) 2014-07-18 2022-03-25 香港中文大学 Tissue methylation pattern analysis in DNA mixtures

Also Published As

Publication number Publication date
EP4063517A1 (en) 2022-09-28
US10689706B2 (en) 2020-06-23
HK1252222A1 (en) 2019-05-24
ES2911613T3 (en) 2022-05-20
EP3325663A4 (en) 2018-07-25
CN108138233B (en) 2022-08-26
US20200270707A1 (en) 2020-08-27
US20170029900A1 (en) 2017-02-02
EP3739061A1 (en) 2020-11-18
CN115287348A (en) 2022-11-04
HUE059407T2 (en) 2022-11-28
PT3739061T (en) 2022-04-05
TWI798718B (en) 2023-04-11
EP3325663B1 (en) 2020-08-19
AU2016295712B2 (en) 2022-06-02
AU2022218555A1 (en) 2022-09-15
TWI732771B (en) 2021-07-11
CA2993588A1 (en) 2017-01-26
TW201718871A (en) 2017-06-01
AU2016295712A1 (en) 2018-02-08
EP3739061B1 (en) 2022-03-23
WO2017012544A1 (en) 2017-01-26
DK3739061T3 (en) 2022-04-19
HK1248283A1 (en) 2018-10-12
CN108138233A (en) 2018-06-08
EP3325663A1 (en) 2018-05-30

Similar Documents

Publication Publication Date Title
US11984195B2 (en) Methylation pattern analysis of tissues in a DNA mixture
TWI732771B (en) Methylation pattern analysis of haplotypes in tissues in a dna mixture
JP6683752B2 (en) Non-invasive determination of fetal or tumor methylome by plasma
JP2022185149A (en) Detecting mutations for cancer screening and fetal analysis
TWI636255B (en) Mutational analysis of plasma dna for cancer detection